【Paddle打比赛】基于PaddleNLP的LAIC2022-小样本多任务学习
【Paddle打比赛】基于PaddleNLP的LAIC2022-小样本多任务学习，本文通过数据处理，AutoTokenizer进行训练，对司法文本进行分类，并抽取命名实体，最终第五名。
·
一、LAIC2022——小样本多任务学习
1.任务介绍
本赛道由中国司法大数据研究院承办。
在司法的各个业务中，存在着丰富的NLP场景，但往往会出现标注样本不足的现象，因此研究小样本场景的模型训练问题就变得非常必要。本赛题发布了关于司法的小样本多任务数据，旨在探索小样本学习最佳模型和在司法上的实践。
数据集下载请访问比赛http://data.court.gov.cn/pages/laic2021.html。该数据只可用于该比赛，未经允许禁止在其他领域中使用。
2.数据介绍
- (1) 本赛题数据来源于法律领域，包括文本分类、命名实体识别任务，总任务数量为10个（分类：案件要素8个、刑档1个，命名实体识别：1个）；
- (2) 训练集：案件要素304条，刑档90条，命名实体识别150条，10个任务共544条数据（包括验证集，不再单独提供验证集，由选手自己切分）；
- (3) 标注样本示例：
2.1 案件要素
案件要素中包含8个二分类子任务，子任务数据格式为json格式，字典包含字段为：
id：文本id。data：文本内容，文书中事实描述部分。label：是否是当前案件要素，是/否。task：任务类型，CLS。
{
"id": 491,
"data": "经审理查明：2017年8月21日13时许，被告人杨信驾驶强制报废的赣0131133变型拖拉机，由北向南行驶至南昌县新莲塔公路曹骆十字路口时，闯红灯继续行驶，将由西向东被害人邓某1驾驶的三轮电动车撞倒，致被害人邓某1受伤。经南昌县公安局交通管理大队道路交通事故认定书认定：被告人杨信负本次事故的主要责任。2017年10月10日，经南昌县公安司法鉴定中心鉴定：被害人邓某1损伤构成重伤二级。2018年2月12日，被害人邓某1死亡。经鉴定被害人邓某1符合因交通事故致颅脑损伤后并衰竭死亡。\n案发后，被告人杨信打电话报警，并随救护车辆送被害人邓某1到医院救治，支付了部分医疗费。2017年10月24日，经传唤，被告人杨信主动向公安机关投案。\n上述事实，被告人在开庭审理过程中亦无异议，且有法医学鉴定书，事故责任认定书，现场勘验、检查笔录，现场照片，事故车辆技术鉴定书，交通事故车辆安全性能鉴定报告书，医疗费票据，预付款收据，归案经过等相关证据证实，足以认定。\n",
"label": ["负"],
"task": "CLS"
}
2.2 刑档
刑档任务数据格式为json格式，字典包含字段：
id：文本id。data：文本内容，文书中事实描述部分。label：文书对应刑档等级，一档/二档/三档。task：任务类型，CLS。
{
"id": 491,
"data": "交通肇事罪河北省本院认为，被告人许祥东违反交通运输管理法规，发生重大交通事故，且肇事后逃逸致人死亡，其行为已构成交通肇事罪，公诉机关指控事实和罪名成立，予以支持。被告人许祥东到案后，如实供述犯罪事实，当庭自愿认罪认罚，可依法从轻处罚。公诉机关提出的对被告人许祥东的量刑意见以及被告人许祥东的辩护人提出的相关辩护意见予以采纳。为维护公共安全，保护公民人身权利，打击刑事犯罪。经审理查明，2019年12月6日6时18分许，在308国道宁晋县处，被告人许祥东无机动车驾驶证驾驶冀E小型轿车由东南向西北行驶时，与由南向西北转弯的田某1驾驶自行车相撞，致田某1及其驾驶的自行车倒地，许祥东驾驶冀E小型轿车逃逸。后由东南向西北行驶的张某驾驶冀A冀A重型仓栅式半挂车又与倒在公路上的田某1及其自行车相撞，造成田某1死亡，三车不同程度损坏的交通事故。许祥东负事故的主要责任。经鉴定，田某1符合车祸致颅脑损伤合并肺脏、肝脏破裂死亡。",
"label": ["三档"],
"task": "CLS"
}
2.3 命名实体识别
命名实体识别任务数据格式为json格式，字典包含字段：
id：案例中文本的唯一标识符。relations：空列表，无用信息。text：文本内容，文书中事实描述部分。entities：句子所包含的实体信息列表。label：实体标签名称。start_offset：实体开始位置下标。end_offset：实体结束位置下标。task：任务类型，NER。
其中 命名实体识别任务中的 label的十种实体类型分别为：
| label | 含义 |
|---|---|
| 11017 | 犯罪嫌疑人情况 |
| 11018 | 被害人 |
| 11019 | 被害人类型 |
| 11020 | 犯罪嫌疑人交通工具 |
| 11021 | 犯罪嫌疑人交通工具情况 |
| 11022 | 被害人交通工具情况 |
| 11023 | 犯罪嫌疑人责任认定 |
| 11024 | 被害人责任认定 |
| 11025 | 事故发生地 |
| 11027 | 被害人交通工具 |
{
"id": 139,
"relations": []
"text": "经审理查明:2016年10月12日10时50分许,被告人昌某无驾驶资格驾驶无号牌机动三轮车沿某政府1门前路段由东向西行驶,行驶过程中该机动三轮车车厢中人员王某2从车上掉落受伤。2016年10月13日,被告人昌某被某政府2处以行政拘留五日、罚款500元的行政处罚;2016年10月16日被害人王某2经医院抢救无效死亡。经某政府2交通警察支队二大队认定,被告人昌某承担本次事故的全部责任。2016年11月30日,被告人昌某接公安机关电话通知后主动到案。",
"entities": [
{"id": 11017, "start_offset": 30, "end_offset": 35, "label": "犯罪嫌疑人情况"},
{"id": 11018, "start_offset": 77, "end_offset": 80, "label": "被害人"},
{"id": 11018, "start_offset": 145, "end_offset": 148, "label": "被害人"},
{"id": 11020, "start_offset": 40, "end_offset": 45, "label": "犯罪嫌疑人交通工具"},
{"id": 11021, "start_offset": 37, "end_offset": 40, "label": "犯罪嫌疑人交通工具情况"},
{"id": 11023, "start_offset": 187, "end_offset": 191, "label": "犯罪嫌疑人责任认定"},
{"id": 11019, "start_offset": 72, "end_offset": 77, "label": "被害人类型"},
{"id": 11025, "start_offset": 46, "end_offset": 54, "label": "事故发生地"}],
"task": "NER"
}
二、数据处理
1.解压缩
!unzip -qoa data/data173002/PaddleNLP-develop.zip
!unzip -qoa data/data173234/1666234655361.zip
2.查看数据
!tree 训练集
训练集
├── ner
├── 案件要素
│ ├── 被害人被后车撞击
│ ├── 被害人闯红灯
│ ├── 被害人为本车人员
│ ├── 交通肇事后逃逸
│ ├── 全部责任
│ ├── 肇事车辆超速行驶
│ ├── 肇事车辆逆行
│ └── 中型客车交通肇事
└── 刑档
1 directory, 10 files
!tree 测试集_选手
测试集_选手
├── ner
├── 案件要素
│ ├── 被害人被后车撞击
│ ├── 被害人闯红灯
│ ├── 被害人为本车人员
│ ├── 交通肇事后逃逸
│ ├── 全部责任
│ ├── 肇事车辆超速行驶
│ ├── 肇事车辆逆行
│ └── 中型客车交通肇事
└── 刑档
1 directory, 10 files
3.数据格式转换
- 转换为doccano数据格式
- 按照 8:2 进行训练集、测试集划分
!python doccano.py \
--splits 0.8 0.2 0
[32m[2022-10-20 21:35:46,089] [ INFO][0m - Converting doccano data...[0m
100%|██████████████████████████████████████| 120/120 [00:00<00:00, 15204.56it/s]
[32m[2022-10-20 21:35:46,099] [ INFO][0m - Adding negative samples for first stage prompt...[0m
100%|██████████████████████████████████████| 120/120 [00:00<00:00, 68702.77it/s]
[32m[2022-10-20 21:35:46,101] [ INFO][0m - Converting doccano data...[0m
100%|████████████████████████████████████████| 30/30 [00:00<00:00, 22052.07it/s]
[32m[2022-10-20 21:35:46,103] [ INFO][0m - Adding negative samples for first stage prompt...[0m
100%|███████████████████████████████████████| 30/30 [00:00<00:00, 114494.19it/s]
[32m[2022-10-20 21:35:46,104] [ INFO][0m - Converting doccano data...[0m
0it [00:00, ?it/s]
[32m[2022-10-20 21:35:46,104] [ INFO][0m - Adding negative samples for first stage prompt...[0m
0it [00:00, ?it/s]
[32m[2022-10-20 21:35:46,116] [ INFO][0m - Save 1188 examples to ./data_save/train.txt.[0m
[32m[2022-10-20 21:35:46,119] [ INFO][0m - Save 300 examples to ./data_save/dev.txt.[0m
[32m[2022-10-20 21:35:46,119] [ INFO][0m - Save 0 examples to ./data_save/test.txt.[0m
[32m[2022-10-20 21:35:46,119] [ INFO][0m - Finished! It takes 0.03 seconds[0m
训练集/案件要素/肇事车辆逆行
[32m[2022-10-20 21:35:46,121] [ INFO][0m - Converting doccano data...[0m
0%| | 0/29 [00:00<?, ?it/s]
[32m[2022-10-20 21:35:46,122] [ INFO][0m - Converting doccano data...[0m
0%| | 0/8 [00:00<?, ?it/s]
[32m[2022-10-20 21:35:46,122] [ INFO][0m - Converting doccano data...[0m
0it [00:00, ?it/s]
[32m[2022-10-20 21:35:46,123] [ INFO][0m - Save 29 examples to ./data_save/train.txt.[0m
[32m[2022-10-20 21:35:46,123] [ INFO][0m - Save 8 examples to ./data_save/dev.txt.[0m
[32m[2022-10-20 21:35:46,123] [ INFO][0m - Save 0 examples to ./data_save/test.txt.[0m
[32m[2022-10-20 21:35:46,123] [ INFO][0m - Finished! It takes 0.00 seconds[0m
训练集/案件要素/中型客车交通肇事
[32m[2022-10-20 21:35:46,124] [ INFO][0m - Converting doccano data...[0m
0%| | 0/30 [00:00<?, ?it/s]
[32m[2022-10-20 21:35:46,125] [ INFO][0m - Converting doccano data...[0m
0%| | 0/8 [00:00<?, ?it/s]
[32m[2022-10-20 21:35:46,125] [ INFO][0m - Converting doccano data...[0m
0it [00:00, ?it/s]
[32m[2022-10-20 21:35:46,126] [ INFO][0m - Save 30 examples to ./data_save/train.txt.[0m
[32m[2022-10-20 21:35:46,126] [ INFO][0m - Save 8 examples to ./data_save/dev.txt.[0m
[32m[2022-10-20 21:35:46,126] [ INFO][0m - Save 0 examples to ./data_save/test.txt.[0m
[32m[2022-10-20 21:35:46,126] [ INFO][0m - Finished! It takes 0.00 seconds[0m
训练集/案件要素/全部责任
[32m[2022-10-20 21:35:46,127] [ INFO][0m - Converting doccano data...[0m
0%| | 0/28 [00:00<?, ?it/s]
[32m[2022-10-20 21:35:46,128] [ INFO][0m - Converting doccano data...[0m
0%| | 0/7 [00:00<?, ?it/s]
[32m[2022-10-20 21:35:46,128] [ INFO][0m - Converting doccano data...[0m
0it [00:00, ?it/s]
[32m[2022-10-20 21:35:46,129] [ INFO][0m - Save 28 examples to ./data_save/train.txt.[0m
[32m[2022-10-20 21:35:46,129] [ INFO][0m - Save 7 examples to ./data_save/dev.txt.[0m
[32m[2022-10-20 21:35:46,129] [ INFO][0m - Save 0 examples to ./data_save/test.txt.[0m
[32m[2022-10-20 21:35:46,129] [ INFO][0m - Finished! It takes 0.00 seconds[0m
训练集/案件要素/交通肇事后逃逸
[32m[2022-10-20 21:35:46,130] [ INFO][0m - Converting doccano data...[0m
0%| | 0/29 [00:00<?, ?it/s]
[32m[2022-10-20 21:35:46,131] [ INFO][0m - Converting doccano data...[0m
0%| | 0/8 [00:00<?, ?it/s]
[32m[2022-10-20 21:35:46,131] [ INFO][0m - Converting doccano data...[0m
0it [00:00, ?it/s]
[32m[2022-10-20 21:35:46,132] [ INFO][0m - Save 29 examples to ./data_save/train.txt.[0m
[32m[2022-10-20 21:35:46,132] [ INFO][0m - Save 8 examples to ./data_save/dev.txt.[0m
[32m[2022-10-20 21:35:46,132] [ INFO][0m - Save 0 examples to ./data_save/test.txt.[0m
[32m[2022-10-20 21:35:46,132] [ INFO][0m - Finished! It takes 0.00 seconds[0m
训练集/案件要素/肇事车辆超速行驶
[32m[2022-10-20 21:35:46,133] [ INFO][0m - Converting doccano data...[0m
0%| | 0/27 [00:00<?, ?it/s]
[32m[2022-10-20 21:35:46,134] [ INFO][0m - Converting doccano data...[0m
0%| | 0/7 [00:00<?, ?it/s]
[32m[2022-10-20 21:35:46,134] [ INFO][0m - Converting doccano data...[0m
0it [00:00, ?it/s]
[32m[2022-10-20 21:35:46,135] [ INFO][0m - Save 27 examples to ./data_save/train.txt.[0m
[32m[2022-10-20 21:35:46,135] [ INFO][0m - Save 7 examples to ./data_save/dev.txt.[0m
[32m[2022-10-20 21:35:46,135] [ INFO][0m - Save 0 examples to ./data_save/test.txt.[0m
[32m[2022-10-20 21:35:46,135] [ INFO][0m - Finished! It takes 0.00 seconds[0m
训练集/案件要素/被害人闯红灯
[32m[2022-10-20 21:35:46,136] [ INFO][0m - Converting doccano data...[0m
0%| | 0/31 [00:00<?, ?it/s]
[32m[2022-10-20 21:35:46,136] [ INFO][0m - Converting doccano data...[0m
0%| | 0/8 [00:00<?, ?it/s]
[32m[2022-10-20 21:35:46,137] [ INFO][0m - Converting doccano data...[0m
0it [00:00, ?it/s]
[32m[2022-10-20 21:35:46,137] [ INFO][0m - Save 31 examples to ./data_save/train.txt.[0m
[32m[2022-10-20 21:35:46,138] [ INFO][0m - Save 8 examples to ./data_save/dev.txt.[0m
[32m[2022-10-20 21:35:46,138] [ INFO][0m - Save 0 examples to ./data_save/test.txt.[0m
[32m[2022-10-20 21:35:46,138] [ INFO][0m - Finished! It takes 0.00 seconds[0m
训练集/案件要素/被害人为本车人员
[32m[2022-10-20 21:35:46,139] [ INFO][0m - Converting doccano data...[0m
0%| | 0/35 [00:00<?, ?it/s]
[32m[2022-10-20 21:35:46,140] [ INFO][0m - Converting doccano data...[0m
0%| | 0/9 [00:00<?, ?it/s]
[32m[2022-10-20 21:35:46,140] [ INFO][0m - Converting doccano data...[0m
0it [00:00, ?it/s]
[32m[2022-10-20 21:35:46,141] [ INFO][0m - Save 35 examples to ./data_save/train.txt.[0m
[32m[2022-10-20 21:35:46,141] [ INFO][0m - Save 9 examples to ./data_save/dev.txt.[0m
[32m[2022-10-20 21:35:46,141] [ INFO][0m - Save 0 examples to ./data_save/test.txt.[0m
[32m[2022-10-20 21:35:46,141] [ INFO][0m - Finished! It takes 0.00 seconds[0m
训练集/案件要素/被害人被后车撞击
[32m[2022-10-20 21:35:46,142] [ INFO][0m - Converting doccano data...[0m
0%| | 0/32 [00:00<?, ?it/s]
[32m[2022-10-20 21:35:46,143] [ INFO][0m - Converting doccano data...[0m
0%| | 0/8 [00:00<?, ?it/s]
[32m[2022-10-20 21:35:46,143] [ INFO][0m - Converting doccano data...[0m
0it [00:00, ?it/s]
[32m[2022-10-20 21:35:46,144] [ INFO][0m - Save 32 examples to ./data_save/train.txt.[0m
[32m[2022-10-20 21:35:46,145] [ INFO][0m - Save 8 examples to ./data_save/dev.txt.[0m
[32m[2022-10-20 21:35:46,145] [ INFO][0m - Save 0 examples to ./data_save/test.txt.[0m
[32m[2022-10-20 21:35:46,145] [ INFO][0m - Finished! It takes 0.00 seconds[0m
训练集/刑档
[32m[2022-10-20 21:35:46,146] [ INFO][0m - Converting doccano data...[0m
0%| | 0/72 [00:00<?, ?it/s]
[32m[2022-10-20 21:35:46,147] [ INFO][0m - Converting doccano data...[0m
0%| | 0/18 [00:00<?, ?it/s]
[32m[2022-10-20 21:35:46,147] [ INFO][0m - Converting doccano data...[0m
0it [00:00, ?it/s]
[32m[2022-10-20 21:35:46,148] [ INFO][0m - Save 72 examples to ./data_save/train.txt.[0m
[32m[2022-10-20 21:35:46,149] [ INFO][0m - Save 18 examples to ./data_save/dev.txt.[0m
[32m[2022-10-20 21:35:46,149] [ INFO][0m - Save 0 examples to ./data_save/test.txt.[0m
[32m[2022-10-20 21:35:46,149] [ INFO][0m - Finished! It takes 0.00 seconds[0m
[0m
!tree data_save
data_save
├── dev.txt
├── test.txt
└── train.txt
0 directories, 3 files
三、模型训练
1.升级paddlenlp
# 升级paddlenlp
!pip install -q paddlenlp==2.3.4
!pip list|grep paddlenlp
paddlenlp 2.3.4
2.模型finetune
%cd ~
!python finetune.py \
--train_path data_save/train.txt \
--dev_path data_save/dev.txt \
--save_dir ./checkpoint \
--learning_rate 5e-5 \
--model uie-base \
--batch_size 40 \
--logging_steps 10 \
--valid_steps 50 \
--num_epochs 40 \
--device gpu
/home/aistudio
[32m[2022-10-20 21:40:07,116] [ INFO][0m - Downloading resource files...[0m
[32m[2022-10-20 21:40:07,122] [ INFO][0m - We are using <class 'paddlenlp.transformers.ernie.tokenizer.ErnieTokenizer'> to load 'uie-base'.[0m
W1020 21:40:07.161936 10428 gpu_resources.cc:61] Please NOTE: device: 0, GPU Compute Capability: 7.0, Driver API Version: 11.2, Runtime API Version: 11.2
W1020 21:40:07.165697 10428 gpu_resources.cc:91] device: 0, cuDNN Version: 8.2.
[32m[2022-10-20 21:40:24,297] [ INFO][0m - global step 10, epoch: 1, loss: 0.00450, speed: 0.72 step/s[0m
[32m[2022-10-20 21:40:36,786] [ INFO][0m - global step 20, epoch: 1, loss: 0.00359, speed: 0.80 step/s[0m
[32m[2022-10-20 21:40:49,101] [ INFO][0m - global step 30, epoch: 1, loss: 0.00326, speed: 0.81 step/s[0m
[32m[2022-10-20 21:41:01,550] [ INFO][0m - global step 40, epoch: 1, loss: 0.00296, speed: 0.80 step/s[0m
[32m[2022-10-20 21:41:14,057] [ INFO][0m - global step 50, epoch: 1, loss: 0.00272, speed: 0.80 step/s[0m
[32m[2022-10-20 21:41:18,235] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_50/tokenizer_config.json[0m
[32m[2022-10-20 21:41:18,236] [ INFO][0m - Special tokens file saved in ./checkpoint/model_50/special_tokens_map.json[0m
[32m[2022-10-20 21:41:35,184] [ INFO][0m - Evaluation precision: 0.87778, recall: 0.60945, F1: 0.71941[0m
[32m[2022-10-20 21:41:35,184] [ INFO][0m - best F1 performence has been updated: 0.00000 --> 0.71941[0m
[32m[2022-10-20 21:41:39,248] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_best/tokenizer_config.json[0m
[32m[2022-10-20 21:41:39,248] [ INFO][0m - Special tokens file saved in ./checkpoint/model_best/special_tokens_map.json[0m
[32m[2022-10-20 21:41:51,387] [ INFO][0m - global step 60, epoch: 1, loss: 0.00253, speed: 0.82 step/s[0m
[32m[2022-10-20 21:42:03,621] [ INFO][0m - global step 70, epoch: 1, loss: 0.00233, speed: 0.82 step/s[0m
[32m[2022-10-20 21:42:16,062] [ INFO][0m - global step 80, epoch: 1, loss: 0.00220, speed: 0.80 step/s[0m
[32m[2022-10-20 21:42:28,615] [ INFO][0m - global step 90, epoch: 1, loss: 0.00210, speed: 0.80 step/s[0m
[32m[2022-10-20 21:42:41,128] [ INFO][0m - global step 100, epoch: 1, loss: 0.00202, speed: 0.80 step/s[0m
[32m[2022-10-20 21:43:01,115] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_100/tokenizer_config.json[0m
[32m[2022-10-20 21:43:01,128] [ INFO][0m - Special tokens file saved in ./checkpoint/model_100/special_tokens_map.json[0m
[32m[2022-10-20 21:43:18,363] [ INFO][0m - Evaluation precision: 0.83678, recall: 0.74156, F1: 0.78630[0m
[32m[2022-10-20 21:43:18,363] [ INFO][0m - best F1 performence has been updated: 0.71941 --> 0.78630[0m
[32m[2022-10-20 21:43:34,566] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_best/tokenizer_config.json[0m
[32m[2022-10-20 21:43:34,584] [ INFO][0m - Special tokens file saved in ./checkpoint/model_best/special_tokens_map.json[0m
[32m[2022-10-20 21:43:47,054] [ INFO][0m - global step 110, epoch: 1, loss: 0.00195, speed: 0.80 step/s[0m
[32m[2022-10-20 21:43:59,372] [ INFO][0m - global step 120, epoch: 1, loss: 0.00189, speed: 0.81 step/s[0m
[32m[2022-10-20 21:44:11,393] [ INFO][0m - global step 130, epoch: 1, loss: 0.00184, speed: 0.83 step/s[0m
[32m[2022-10-20 21:44:23,091] [ INFO][0m - global step 140, epoch: 2, loss: 0.00178, speed: 0.85 step/s[0m
[32m[2022-10-20 21:44:35,283] [ INFO][0m - global step 150, epoch: 2, loss: 0.00172, speed: 0.82 step/s[0m
[32m[2022-10-20 21:44:54,250] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_150/tokenizer_config.json[0m
[32m[2022-10-20 21:44:54,379] [ INFO][0m - Special tokens file saved in ./checkpoint/model_150/special_tokens_map.json[0m
[32m[2022-10-20 21:45:11,209] [ INFO][0m - Evaluation precision: 0.85896, recall: 0.78110, F1: 0.81818[0m
[32m[2022-10-20 21:45:11,210] [ INFO][0m - best F1 performence has been updated: 0.78630 --> 0.81818[0m
[32m[2022-10-20 21:45:16,040] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_best/tokenizer_config.json[0m
[32m[2022-10-20 21:45:16,040] [ INFO][0m - Special tokens file saved in ./checkpoint/model_best/special_tokens_map.json[0m
[32m[2022-10-20 21:45:28,343] [ INFO][0m - global step 160, epoch: 2, loss: 0.00165, speed: 0.81 step/s[0m
[32m[2022-10-20 21:45:40,450] [ INFO][0m - global step 170, epoch: 2, loss: 0.00160, speed: 0.83 step/s[0m
[32m[2022-10-20 21:45:52,626] [ INFO][0m - global step 180, epoch: 2, loss: 0.00156, speed: 0.82 step/s[0m
[32m[2022-10-20 21:46:04,713] [ INFO][0m - global step 190, epoch: 2, loss: 0.00152, speed: 0.83 step/s[0m
[32m[2022-10-20 21:46:16,872] [ INFO][0m - global step 200, epoch: 2, loss: 0.00149, speed: 0.82 step/s[0m
[32m[2022-10-20 21:46:21,323] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_200/tokenizer_config.json[0m
[32m[2022-10-20 21:46:21,323] [ INFO][0m - Special tokens file saved in ./checkpoint/model_200/special_tokens_map.json[0m
[32m[2022-10-20 21:46:39,060] [ INFO][0m - Evaluation precision: 0.83247, recall: 0.77628, F1: 0.80339[0m
[32m[2022-10-20 21:46:51,257] [ INFO][0m - global step 210, epoch: 2, loss: 0.00145, speed: 0.82 step/s[0m
[32m[2022-10-20 21:47:03,591] [ INFO][0m - global step 220, epoch: 2, loss: 0.00141, speed: 0.81 step/s[0m
[32m[2022-10-20 21:47:15,894] [ INFO][0m - global step 230, epoch: 2, loss: 0.00138, speed: 0.81 step/s[0m
[32m[2022-10-20 21:47:27,997] [ INFO][0m - global step 240, epoch: 2, loss: 0.00134, speed: 0.83 step/s[0m
[32m[2022-10-20 21:47:40,374] [ INFO][0m - global step 250, epoch: 2, loss: 0.00132, speed: 0.81 step/s[0m
[32m[2022-10-20 21:47:44,840] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_250/tokenizer_config.json[0m
[32m[2022-10-20 21:47:44,841] [ INFO][0m - Special tokens file saved in ./checkpoint/model_250/special_tokens_map.json[0m
[32m[2022-10-20 21:48:01,845] [ INFO][0m - Evaluation precision: 0.88840, recall: 0.78303, F1: 0.83239[0m
[32m[2022-10-20 21:48:01,845] [ INFO][0m - best F1 performence has been updated: 0.81818 --> 0.83239[0m
[32m[2022-10-20 21:48:06,862] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_best/tokenizer_config.json[0m
[32m[2022-10-20 21:48:06,862] [ INFO][0m - Special tokens file saved in ./checkpoint/model_best/special_tokens_map.json[0m
[32m[2022-10-20 21:48:18,952] [ INFO][0m - global step 260, epoch: 2, loss: 0.00129, speed: 0.83 step/s[0m
[32m[2022-10-20 21:48:29,892] [ INFO][0m - global step 270, epoch: 3, loss: 0.00127, speed: 0.91 step/s[0m
[32m[2022-10-20 21:48:41,772] [ INFO][0m - global step 280, epoch: 3, loss: 0.00123, speed: 0.84 step/s[0m
[32m[2022-10-20 21:48:54,294] [ INFO][0m - global step 290, epoch: 3, loss: 0.00120, speed: 0.80 step/s[0m
[32m[2022-10-20 21:49:06,182] [ INFO][0m - global step 300, epoch: 3, loss: 0.00118, speed: 0.84 step/s[0m
[32m[2022-10-20 21:49:07,563] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_300/tokenizer_config.json[0m
[32m[2022-10-20 21:49:07,564] [ INFO][0m - Special tokens file saved in ./checkpoint/model_300/special_tokens_map.json[0m
[32m[2022-10-20 21:49:24,397] [ INFO][0m - Evaluation precision: 0.84065, recall: 0.79364, F1: 0.81647[0m
[32m[2022-10-20 21:49:36,708] [ INFO][0m - global step 310, epoch: 3, loss: 0.00116, speed: 0.81 step/s[0m
[32m[2022-10-20 21:49:48,871] [ INFO][0m - global step 320, epoch: 3, loss: 0.00113, speed: 0.82 step/s[0m
[32m[2022-10-20 21:50:01,308] [ INFO][0m - global step 330, epoch: 3, loss: 0.00111, speed: 0.80 step/s[0m
[32m[2022-10-20 21:50:13,553] [ INFO][0m - global step 340, epoch: 3, loss: 0.00109, speed: 0.82 step/s[0m
[32m[2022-10-20 21:50:25,837] [ INFO][0m - global step 350, epoch: 3, loss: 0.00108, speed: 0.81 step/s[0m
[32m[2022-10-20 21:50:27,171] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_350/tokenizer_config.json[0m
[32m[2022-10-20 21:50:27,171] [ INFO][0m - Special tokens file saved in ./checkpoint/model_350/special_tokens_map.json[0m
[32m[2022-10-20 21:50:43,769] [ INFO][0m - Evaluation precision: 0.86680, recall: 0.82835, F1: 0.84714[0m
[32m[2022-10-20 21:50:43,769] [ INFO][0m - best F1 performence has been updated: 0.83239 --> 0.84714[0m
[32m[2022-10-20 21:50:48,466] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_best/tokenizer_config.json[0m
[32m[2022-10-20 21:50:48,466] [ INFO][0m - Special tokens file saved in ./checkpoint/model_best/special_tokens_map.json[0m
[32m[2022-10-20 21:51:00,793] [ INFO][0m - global step 360, epoch: 3, loss: 0.00106, speed: 0.81 step/s[0m
[32m[2022-10-20 21:51:13,089] [ INFO][0m - global step 370, epoch: 3, loss: 0.00104, speed: 0.81 step/s[0m
[32m[2022-10-20 21:51:25,395] [ INFO][0m - global step 380, epoch: 3, loss: 0.00103, speed: 0.81 step/s[0m
[32m[2022-10-20 21:51:37,530] [ INFO][0m - global step 390, epoch: 3, loss: 0.00101, speed: 0.82 step/s[0m
[32m[2022-10-20 21:51:49,229] [ INFO][0m - global step 400, epoch: 3, loss: 0.00099, speed: 0.85 step/s[0m
[32m[2022-10-20 21:51:50,541] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_400/tokenizer_config.json[0m
[32m[2022-10-20 21:51:50,542] [ INFO][0m - Special tokens file saved in ./checkpoint/model_400/special_tokens_map.json[0m
[32m[2022-10-20 21:52:07,312] [ INFO][0m - Evaluation precision: 0.87639, recall: 0.83414, F1: 0.85474[0m
[32m[2022-10-20 21:52:07,312] [ INFO][0m - best F1 performence has been updated: 0.84714 --> 0.85474[0m
[32m[2022-10-20 21:52:11,868] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_best/tokenizer_config.json[0m
[32m[2022-10-20 21:52:11,868] [ INFO][0m - Special tokens file saved in ./checkpoint/model_best/special_tokens_map.json[0m
[32m[2022-10-20 21:52:23,602] [ INFO][0m - global step 410, epoch: 4, loss: 0.00097, speed: 0.85 step/s[0m
[32m[2022-10-20 21:52:35,855] [ INFO][0m - global step 420, epoch: 4, loss: 0.00096, speed: 0.82 step/s[0m
[32m[2022-10-20 21:52:48,174] [ INFO][0m - global step 430, epoch: 4, loss: 0.00094, speed: 0.81 step/s[0m
[32m[2022-10-20 21:53:00,544] [ INFO][0m - global step 440, epoch: 4, loss: 0.00093, speed: 0.81 step/s[0m
[32m[2022-10-20 21:53:13,104] [ INFO][0m - global step 450, epoch: 4, loss: 0.00091, speed: 0.80 step/s[0m
[32m[2022-10-20 21:53:14,474] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_450/tokenizer_config.json[0m
[32m[2022-10-20 21:53:14,475] [ INFO][0m - Special tokens file saved in ./checkpoint/model_450/special_tokens_map.json[0m
[32m[2022-10-20 21:53:30,915] [ INFO][0m - Evaluation precision: 0.86263, recall: 0.82353, F1: 0.84262[0m
[32m[2022-10-20 21:53:42,729] [ INFO][0m - global step 460, epoch: 4, loss: 0.00090, speed: 0.85 step/s[0m
[32m[2022-10-20 21:53:54,816] [ INFO][0m - global step 470, epoch: 4, loss: 0.00088, speed: 0.83 step/s[0m
[32m[2022-10-20 21:54:07,106] [ INFO][0m - global step 480, epoch: 4, loss: 0.00087, speed: 0.81 step/s[0m
[32m[2022-10-20 21:54:19,535] [ INFO][0m - global step 490, epoch: 4, loss: 0.00086, speed: 0.80 step/s[0m
[32m[2022-10-20 21:54:31,569] [ INFO][0m - global step 500, epoch: 4, loss: 0.00085, speed: 0.83 step/s[0m
[32m[2022-10-20 21:54:32,902] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_500/tokenizer_config.json[0m
[32m[2022-10-20 21:54:32,902] [ INFO][0m - Special tokens file saved in ./checkpoint/model_500/special_tokens_map.json[0m
[32m[2022-10-20 21:54:49,895] [ INFO][0m - Evaluation precision: 0.88163, recall: 0.83317, F1: 0.85672[0m
[32m[2022-10-20 21:54:49,895] [ INFO][0m - best F1 performence has been updated: 0.85474 --> 0.85672[0m
[32m[2022-10-20 21:54:54,632] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_best/tokenizer_config.json[0m
[32m[2022-10-20 21:54:54,632] [ INFO][0m - Special tokens file saved in ./checkpoint/model_best/special_tokens_map.json[0m
[32m[2022-10-20 21:55:06,912] [ INFO][0m - global step 510, epoch: 4, loss: 0.00083, speed: 0.81 step/s[0m
[32m[2022-10-20 21:55:19,302] [ INFO][0m - global step 520, epoch: 4, loss: 0.00083, speed: 0.81 step/s[0m
[32m[2022-10-20 21:55:31,588] [ INFO][0m - global step 530, epoch: 4, loss: 0.00082, speed: 0.81 step/s[0m
[32m[2022-10-20 21:55:42,947] [ INFO][0m - global step 540, epoch: 5, loss: 0.00081, speed: 0.88 step/s[0m
[32m[2022-10-20 21:55:55,285] [ INFO][0m - global step 550, epoch: 5, loss: 0.00079, speed: 0.81 step/s[0m
[32m[2022-10-20 21:55:56,623] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_550/tokenizer_config.json[0m
[32m[2022-10-20 21:55:56,623] [ INFO][0m - Special tokens file saved in ./checkpoint/model_550/special_tokens_map.json[0m
[32m[2022-10-20 21:56:15,152] [ INFO][0m - Evaluation precision: 0.85513, recall: 0.81967, F1: 0.83703[0m
[32m[2022-10-20 21:56:27,563] [ INFO][0m - global step 560, epoch: 5, loss: 0.00078, speed: 0.81 step/s[0m
[32m[2022-10-20 21:56:40,082] [ INFO][0m - global step 570, epoch: 5, loss: 0.00077, speed: 0.80 step/s[0m
[32m[2022-10-20 21:56:52,644] [ INFO][0m - global step 580, epoch: 5, loss: 0.00076, speed: 0.80 step/s[0m
[32m[2022-10-20 21:57:04,831] [ INFO][0m - global step 590, epoch: 5, loss: 0.00075, speed: 0.82 step/s[0m
[32m[2022-10-20 21:57:17,119] [ INFO][0m - global step 600, epoch: 5, loss: 0.00075, speed: 0.81 step/s[0m
[32m[2022-10-20 21:57:18,474] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_600/tokenizer_config.json[0m
[32m[2022-10-20 21:57:18,474] [ INFO][0m - Special tokens file saved in ./checkpoint/model_600/special_tokens_map.json[0m
[32m[2022-10-20 21:57:36,861] [ INFO][0m - Evaluation precision: 0.84429, recall: 0.84185, F1: 0.84307[0m
[32m[2022-10-20 21:57:48,999] [ INFO][0m - global step 610, epoch: 5, loss: 0.00074, speed: 0.82 step/s[0m
[32m[2022-10-20 21:58:01,156] [ INFO][0m - global step 620, epoch: 5, loss: 0.00073, speed: 0.82 step/s[0m
[32m[2022-10-20 21:58:13,390] [ INFO][0m - global step 630, epoch: 5, loss: 0.00073, speed: 0.82 step/s[0m
[32m[2022-10-20 21:58:25,739] [ INFO][0m - global step 640, epoch: 5, loss: 0.00072, speed: 0.81 step/s[0m
[32m[2022-10-20 21:58:38,418] [ INFO][0m - global step 650, epoch: 5, loss: 0.00071, speed: 0.79 step/s[0m
[32m[2022-10-20 21:58:39,816] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_650/tokenizer_config.json[0m
[32m[2022-10-20 21:58:39,817] [ INFO][0m - Special tokens file saved in ./checkpoint/model_650/special_tokens_map.json[0m
[32m[2022-10-20 21:58:57,101] [ INFO][0m - Evaluation precision: 0.90000, recall: 0.84185, F1: 0.86996[0m
[32m[2022-10-20 21:58:57,101] [ INFO][0m - best F1 performence has been updated: 0.85672 --> 0.86996[0m
[32m[2022-10-20 21:59:01,553] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_best/tokenizer_config.json[0m
[32m[2022-10-20 21:59:01,553] [ INFO][0m - Special tokens file saved in ./checkpoint/model_best/special_tokens_map.json[0m
[32m[2022-10-20 21:59:13,853] [ INFO][0m - global step 660, epoch: 5, loss: 0.00070, speed: 0.81 step/s[0m
[32m[2022-10-20 21:59:24,362] [ INFO][0m - global step 670, epoch: 5, loss: 0.00070, speed: 0.95 step/s[0m
[32m[2022-10-20 21:59:37,298] [ INFO][0m - global step 680, epoch: 6, loss: 0.00069, speed: 0.77 step/s[0m
[32m[2022-10-20 21:59:49,620] [ INFO][0m - global step 690, epoch: 6, loss: 0.00068, speed: 0.81 step/s[0m
[32m[2022-10-20 22:00:01,943] [ INFO][0m - global step 700, epoch: 6, loss: 0.00067, speed: 0.81 step/s[0m
[32m[2022-10-20 22:00:03,263] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_700/tokenizer_config.json[0m
[32m[2022-10-20 22:00:03,263] [ INFO][0m - Special tokens file saved in ./checkpoint/model_700/special_tokens_map.json[0m
[32m[2022-10-20 22:00:20,056] [ INFO][0m - Evaluation precision: 0.88666, recall: 0.85246, F1: 0.86922[0m
[32m[2022-10-20 22:00:32,422] [ INFO][0m - global step 710, epoch: 6, loss: 0.00066, speed: 0.81 step/s[0m
[32m[2022-10-20 22:00:44,566] [ INFO][0m - global step 720, epoch: 6, loss: 0.00065, speed: 0.82 step/s[0m
[32m[2022-10-20 22:00:56,856] [ INFO][0m - global step 730, epoch: 6, loss: 0.00065, speed: 0.81 step/s[0m
[32m[2022-10-20 22:01:09,042] [ INFO][0m - global step 740, epoch: 6, loss: 0.00064, speed: 0.82 step/s[0m
[32m[2022-10-20 22:01:21,301] [ INFO][0m - global step 750, epoch: 6, loss: 0.00063, speed: 0.82 step/s[0m
[32m[2022-10-20 22:01:22,627] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_750/tokenizer_config.json[0m
[32m[2022-10-20 22:01:22,628] [ INFO][0m - Special tokens file saved in ./checkpoint/model_750/special_tokens_map.json[0m
[32m[2022-10-20 22:01:39,348] [ INFO][0m - Evaluation precision: 0.90342, recall: 0.86596, F1: 0.88429[0m
[32m[2022-10-20 22:01:39,348] [ INFO][0m - best F1 performence has been updated: 0.86996 --> 0.88429[0m
[32m[2022-10-20 22:01:43,756] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_best/tokenizer_config.json[0m
[32m[2022-10-20 22:01:43,756] [ INFO][0m - Special tokens file saved in ./checkpoint/model_best/special_tokens_map.json[0m
[32m[2022-10-20 22:01:56,164] [ INFO][0m - global step 760, epoch: 6, loss: 0.00063, speed: 0.81 step/s[0m
[32m[2022-10-20 22:02:08,431] [ INFO][0m - global step 770, epoch: 6, loss: 0.00062, speed: 0.82 step/s[0m
[32m[2022-10-20 22:02:20,722] [ INFO][0m - global step 780, epoch: 6, loss: 0.00062, speed: 0.81 step/s[0m
[32m[2022-10-20 22:02:32,838] [ INFO][0m - global step 790, epoch: 6, loss: 0.00061, speed: 0.83 step/s[0m
[32m[2022-10-20 22:02:44,744] [ INFO][0m - global step 800, epoch: 6, loss: 0.00060, speed: 0.84 step/s[0m
[32m[2022-10-20 22:02:46,078] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_800/tokenizer_config.json[0m
[32m[2022-10-20 22:02:46,078] [ INFO][0m - Special tokens file saved in ./checkpoint/model_800/special_tokens_map.json[0m
[32m[2022-10-20 22:03:02,682] [ INFO][0m - Evaluation precision: 0.90010, recall: 0.86017, F1: 0.87968[0m
[32m[2022-10-20 22:03:14,345] [ INFO][0m - global step 810, epoch: 7, loss: 0.00060, speed: 0.86 step/s[0m
[32m[2022-10-20 22:03:26,448] [ INFO][0m - global step 820, epoch: 7, loss: 0.00059, speed: 0.83 step/s[0m
[32m[2022-10-20 22:03:38,825] [ INFO][0m - global step 830, epoch: 7, loss: 0.00059, speed: 0.81 step/s[0m
[32m[2022-10-20 22:03:51,058] [ INFO][0m - global step 840, epoch: 7, loss: 0.00058, speed: 0.82 step/s[0m
[32m[2022-10-20 22:04:03,216] [ INFO][0m - global step 850, epoch: 7, loss: 0.00058, speed: 0.82 step/s[0m
[32m[2022-10-20 22:04:04,545] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_850/tokenizer_config.json[0m
[32m[2022-10-20 22:04:04,546] [ INFO][0m - Special tokens file saved in ./checkpoint/model_850/special_tokens_map.json[0m
[32m[2022-10-20 22:04:22,201] [ INFO][0m - Evaluation precision: 0.90615, recall: 0.83799, F1: 0.87074[0m
[32m[2022-10-20 22:04:34,345] [ INFO][0m - global step 860, epoch: 7, loss: 0.00057, speed: 0.82 step/s[0m
[32m[2022-10-20 22:04:46,677] [ INFO][0m - global step 870, epoch: 7, loss: 0.00056, speed: 0.81 step/s[0m
[32m[2022-10-20 22:04:59,021] [ INFO][0m - global step 880, epoch: 7, loss: 0.00056, speed: 0.81 step/s[0m
[32m[2022-10-20 22:05:11,537] [ INFO][0m - global step 890, epoch: 7, loss: 0.00055, speed: 0.80 step/s[0m
[32m[2022-10-20 22:05:23,897] [ INFO][0m - global step 900, epoch: 7, loss: 0.00055, speed: 0.81 step/s[0m
[32m[2022-10-20 22:05:25,129] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_900/tokenizer_config.json[0m
[32m[2022-10-20 22:05:25,129] [ INFO][0m - Special tokens file saved in ./checkpoint/model_900/special_tokens_map.json[0m
[32m[2022-10-20 22:05:41,809] [ INFO][0m - Evaluation precision: 0.89759, recall: 0.86210, F1: 0.87949[0m
[32m[2022-10-20 22:05:54,153] [ INFO][0m - global step 910, epoch: 7, loss: 0.00054, speed: 0.81 step/s[0m
[32m[2022-10-20 22:06:06,433] [ INFO][0m - global step 920, epoch: 7, loss: 0.00054, speed: 0.81 step/s[0m
[32m[2022-10-20 22:06:18,671] [ INFO][0m - global step 930, epoch: 7, loss: 0.00053, speed: 0.82 step/s[0m
[32m[2022-10-20 22:06:29,933] [ INFO][0m - global step 940, epoch: 8, loss: 0.00053, speed: 0.89 step/s[0m
[32m[2022-10-20 22:06:42,204] [ INFO][0m - global step 950, epoch: 8, loss: 0.00052, speed: 0.81 step/s[0m
[32m[2022-10-20 22:06:43,428] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_950/tokenizer_config.json[0m
[32m[2022-10-20 22:06:43,428] [ INFO][0m - Special tokens file saved in ./checkpoint/model_950/special_tokens_map.json[0m
[32m[2022-10-20 22:07:00,338] [ INFO][0m - Evaluation precision: 0.90909, recall: 0.85824, F1: 0.88294[0m
[32m[2022-10-20 22:07:12,614] [ INFO][0m - global step 960, epoch: 8, loss: 0.00052, speed: 0.81 step/s[0m
[32m[2022-10-20 22:07:24,885] [ INFO][0m - global step 970, epoch: 8, loss: 0.00051, speed: 0.81 step/s[0m
[32m[2022-10-20 22:07:37,249] [ INFO][0m - global step 980, epoch: 8, loss: 0.00051, speed: 0.81 step/s[0m
[32m[2022-10-20 22:07:49,686] [ INFO][0m - global step 990, epoch: 8, loss: 0.00050, speed: 0.80 step/s[0m
[32m[2022-10-20 22:08:02,048] [ INFO][0m - global step 1000, epoch: 8, loss: 0.00050, speed: 0.81 step/s[0m
[32m[2022-10-20 22:08:03,307] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_1000/tokenizer_config.json[0m
[32m[2022-10-20 22:08:03,307] [ INFO][0m - Special tokens file saved in ./checkpoint/model_1000/special_tokens_map.json[0m
[32m[2022-10-20 22:08:20,118] [ INFO][0m - Evaluation precision: 0.90080, recall: 0.86692, F1: 0.88354[0m
[32m[2022-10-20 22:08:32,364] [ INFO][0m - global step 1010, epoch: 8, loss: 0.00049, speed: 0.82 step/s[0m
[32m[2022-10-20 22:08:44,735] [ INFO][0m - global step 1020, epoch: 8, loss: 0.00049, speed: 0.81 step/s[0m
[32m[2022-10-20 22:08:56,906] [ INFO][0m - global step 1030, epoch: 8, loss: 0.00049, speed: 0.82 step/s[0m
[32m[2022-10-20 22:09:09,285] [ INFO][0m - global step 1040, epoch: 8, loss: 0.00048, speed: 0.81 step/s[0m
[32m[2022-10-20 22:09:21,440] [ INFO][0m - global step 1050, epoch: 8, loss: 0.00048, speed: 0.82 step/s[0m
[32m[2022-10-20 22:09:22,658] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_1050/tokenizer_config.json[0m
[32m[2022-10-20 22:09:22,659] [ INFO][0m - Special tokens file saved in ./checkpoint/model_1050/special_tokens_map.json[0m
[32m[2022-10-20 22:09:39,620] [ INFO][0m - Evaluation precision: 0.89819, recall: 0.85921, F1: 0.87827[0m
[32m[2022-10-20 22:09:51,971] [ INFO][0m - global step 1060, epoch: 8, loss: 0.00048, speed: 0.81 step/s[0m
[32m[2022-10-20 22:10:03,630] [ INFO][0m - global step 1070, epoch: 8, loss: 0.00047, speed: 0.86 step/s[0m
[32m[2022-10-20 22:10:15,821] [ INFO][0m - global step 1080, epoch: 9, loss: 0.00047, speed: 0.82 step/s[0m
[32m[2022-10-20 22:10:28,134] [ INFO][0m - global step 1090, epoch: 9, loss: 0.00046, speed: 0.81 step/s[0m
[32m[2022-10-20 22:10:40,217] [ INFO][0m - global step 1100, epoch: 9, loss: 0.00046, speed: 0.83 step/s[0m
[32m[2022-10-20 22:10:41,568] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_1100/tokenizer_config.json[0m
[32m[2022-10-20 22:10:41,568] [ INFO][0m - Special tokens file saved in ./checkpoint/model_1100/special_tokens_map.json[0m
[32m[2022-10-20 22:10:58,370] [ INFO][0m - Evaluation precision: 0.91203, recall: 0.86982, F1: 0.89042[0m
[32m[2022-10-20 22:10:58,370] [ INFO][0m - best F1 performence has been updated: 0.88429 --> 0.89042[0m
[32m[2022-10-20 22:11:03,049] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_best/tokenizer_config.json[0m
[32m[2022-10-20 22:11:03,049] [ INFO][0m - Special tokens file saved in ./checkpoint/model_best/special_tokens_map.json[0m
[32m[2022-10-20 22:11:14,680] [ INFO][0m - global step 1110, epoch: 9, loss: 0.00046, speed: 0.86 step/s[0m
[32m[2022-10-20 22:11:26,715] [ INFO][0m - global step 1120, epoch: 9, loss: 0.00045, speed: 0.83 step/s[0m
[32m[2022-10-20 22:11:39,106] [ INFO][0m - global step 1130, epoch: 9, loss: 0.00045, speed: 0.81 step/s[0m
[32m[2022-10-20 22:11:50,962] [ INFO][0m - global step 1140, epoch: 9, loss: 0.00045, speed: 0.84 step/s[0m
[32m[2022-10-20 22:12:03,208] [ INFO][0m - global step 1150, epoch: 9, loss: 0.00044, speed: 0.82 step/s[0m
[32m[2022-10-20 22:12:04,540] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_1150/tokenizer_config.json[0m
[32m[2022-10-20 22:12:04,540] [ INFO][0m - Special tokens file saved in ./checkpoint/model_1150/special_tokens_map.json[0m
[32m[2022-10-20 22:12:21,594] [ INFO][0m - Evaluation precision: 0.89780, recall: 0.86403, F1: 0.88059[0m
[32m[2022-10-20 22:12:33,860] [ INFO][0m - global step 1160, epoch: 9, loss: 0.00044, speed: 0.82 step/s[0m
[32m[2022-10-20 22:12:46,211] [ INFO][0m - global step 1170, epoch: 9, loss: 0.00044, speed: 0.81 step/s[0m
[32m[2022-10-20 22:12:58,541] [ INFO][0m - global step 1180, epoch: 9, loss: 0.00043, speed: 0.81 step/s[0m
[32m[2022-10-20 22:13:10,917] [ INFO][0m - global step 1190, epoch: 9, loss: 0.00043, speed: 0.81 step/s[0m
[32m[2022-10-20 22:13:23,168] [ INFO][0m - global step 1200, epoch: 9, loss: 0.00043, speed: 0.82 step/s[0m
[32m[2022-10-20 22:13:24,405] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_1200/tokenizer_config.json[0m
[32m[2022-10-20 22:13:24,406] [ INFO][0m - Special tokens file saved in ./checkpoint/model_1200/special_tokens_map.json[0m
[32m[2022-10-20 22:13:41,594] [ INFO][0m - Evaluation precision: 0.90410, recall: 0.87271, F1: 0.88813[0m
[32m[2022-10-20 22:13:52,766] [ INFO][0m - global step 1210, epoch: 10, loss: 0.00042, speed: 0.90 step/s[0m
[32m[2022-10-20 22:14:05,036] [ INFO][0m - global step 1220, epoch: 10, loss: 0.00042, speed: 0.81 step/s[0m
[32m[2022-10-20 22:14:16,962] [ INFO][0m - global step 1230, epoch: 10, loss: 0.00042, speed: 0.84 step/s[0m
[32m[2022-10-20 22:14:28,926] [ INFO][0m - global step 1240, epoch: 10, loss: 0.00041, speed: 0.84 step/s[0m
[32m[2022-10-20 22:14:41,015] [ INFO][0m - global step 1250, epoch: 10, loss: 0.00041, speed: 0.83 step/s[0m
[32m[2022-10-20 22:14:42,277] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_1250/tokenizer_config.json[0m
[32m[2022-10-20 22:14:42,277] [ INFO][0m - Special tokens file saved in ./checkpoint/model_1250/special_tokens_map.json[0m
[32m[2022-10-20 22:14:59,440] [ INFO][0m - Evaluation precision: 0.91009, recall: 0.87850, F1: 0.89401[0m
[32m[2022-10-20 22:14:59,440] [ INFO][0m - best F1 performence has been updated: 0.89042 --> 0.89401[0m
[32m[2022-10-20 22:15:03,833] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_best/tokenizer_config.json[0m
[32m[2022-10-20 22:15:03,833] [ INFO][0m - Special tokens file saved in ./checkpoint/model_best/special_tokens_map.json[0m
[32m[2022-10-20 22:15:15,869] [ INFO][0m - global step 1260, epoch: 10, loss: 0.00041, speed: 0.83 step/s[0m
[32m[2022-10-20 22:15:28,639] [ INFO][0m - global step 1270, epoch: 10, loss: 0.00040, speed: 0.78 step/s[0m
[32m[2022-10-20 22:15:41,369] [ INFO][0m - global step 1280, epoch: 10, loss: 0.00040, speed: 0.79 step/s[0m
[32m[2022-10-20 22:15:53,338] [ INFO][0m - global step 1290, epoch: 10, loss: 0.00040, speed: 0.84 step/s[0m
[32m[2022-10-20 22:16:05,900] [ INFO][0m - global step 1300, epoch: 10, loss: 0.00040, speed: 0.80 step/s[0m
[32m[2022-10-20 22:16:07,415] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_1300/tokenizer_config.json[0m
[32m[2022-10-20 22:16:07,416] [ INFO][0m - Special tokens file saved in ./checkpoint/model_1300/special_tokens_map.json[0m
[32m[2022-10-20 22:16:23,988] [ INFO][0m - Evaluation precision: 0.90746, recall: 0.87946, F1: 0.89324[0m
[32m[2022-10-20 22:16:36,524] [ INFO][0m - global step 1310, epoch: 10, loss: 0.00039, speed: 0.80 step/s[0m
[32m[2022-10-20 22:16:48,514] [ INFO][0m - global step 1320, epoch: 10, loss: 0.00039, speed: 0.83 step/s[0m
[32m[2022-10-20 22:17:01,259] [ INFO][0m - global step 1330, epoch: 10, loss: 0.00039, speed: 0.78 step/s[0m
[32m[2022-10-20 22:17:11,931] [ INFO][0m - global step 1340, epoch: 10, loss: 0.00038, speed: 0.94 step/s[0m
[32m[2022-10-20 22:17:25,440] [ INFO][0m - global step 1350, epoch: 11, loss: 0.00038, speed: 0.74 step/s[0m
[32m[2022-10-20 22:17:26,700] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_1350/tokenizer_config.json[0m
[32m[2022-10-20 22:17:26,700] [ INFO][0m - Special tokens file saved in ./checkpoint/model_1350/special_tokens_map.json[0m
[32m[2022-10-20 22:17:41,754] [ INFO][0m - Evaluation precision: 0.90059, recall: 0.88235, F1: 0.89138[0m
[32m[2022-10-20 22:17:54,420] [ INFO][0m - global step 1360, epoch: 11, loss: 0.00038, speed: 0.79 step/s[0m
[32m[2022-10-20 22:18:07,000] [ INFO][0m - global step 1370, epoch: 11, loss: 0.00038, speed: 0.79 step/s[0m
[32m[2022-10-20 22:18:19,550] [ INFO][0m - global step 1380, epoch: 11, loss: 0.00037, speed: 0.80 step/s[0m
[32m[2022-10-20 22:18:31,732] [ INFO][0m - global step 1390, epoch: 11, loss: 0.00037, speed: 0.82 step/s[0m
[32m[2022-10-20 22:18:44,004] [ INFO][0m - global step 1400, epoch: 11, loss: 0.00037, speed: 0.81 step/s[0m
[32m[2022-10-20 22:18:45,252] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_1400/tokenizer_config.json[0m
[32m[2022-10-20 22:18:45,253] [ INFO][0m - Special tokens file saved in ./checkpoint/model_1400/special_tokens_map.json[0m
[32m[2022-10-20 22:19:01,787] [ INFO][0m - Evaluation precision: 0.90819, recall: 0.88717, F1: 0.89756[0m
[32m[2022-10-20 22:19:01,788] [ INFO][0m - best F1 performence has been updated: 0.89401 --> 0.89756[0m
[32m[2022-10-20 22:19:06,209] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_best/tokenizer_config.json[0m
[32m[2022-10-20 22:19:06,210] [ INFO][0m - Special tokens file saved in ./checkpoint/model_best/special_tokens_map.json[0m
[32m[2022-10-20 22:19:18,629] [ INFO][0m - global step 1410, epoch: 11, loss: 0.00037, speed: 0.81 step/s[0m
[32m[2022-10-20 22:19:30,948] [ INFO][0m - global step 1420, epoch: 11, loss: 0.00036, speed: 0.81 step/s[0m
[32m[2022-10-20 22:19:42,921] [ INFO][0m - global step 1430, epoch: 11, loss: 0.00036, speed: 0.84 step/s[0m
[32m[2022-10-20 22:19:55,447] [ INFO][0m - global step 1440, epoch: 11, loss: 0.00036, speed: 0.80 step/s[0m
[32m[2022-10-20 22:20:07,882] [ INFO][0m - global step 1450, epoch: 11, loss: 0.00036, speed: 0.80 step/s[0m
[32m[2022-10-20 22:20:09,133] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_1450/tokenizer_config.json[0m
[32m[2022-10-20 22:20:09,134] [ INFO][0m - Special tokens file saved in ./checkpoint/model_1450/special_tokens_map.json[0m
[32m[2022-10-20 22:20:26,827] [ INFO][0m - Evaluation precision: 0.91089, recall: 0.88717, F1: 0.89888[0m
[32m[2022-10-20 22:20:26,828] [ INFO][0m - best F1 performence has been updated: 0.89756 --> 0.89888[0m
[32m[2022-10-20 22:20:31,208] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_best/tokenizer_config.json[0m
[32m[2022-10-20 22:20:31,208] [ INFO][0m - Special tokens file saved in ./checkpoint/model_best/special_tokens_map.json[0m
[32m[2022-10-20 22:20:43,640] [ INFO][0m - global step 1460, epoch: 11, loss: 0.00035, speed: 0.81 step/s[0m
[32m[2022-10-20 22:20:55,552] [ INFO][0m - global step 1470, epoch: 11, loss: 0.00035, speed: 0.84 step/s[0m
[32m[2022-10-20 22:21:08,076] [ INFO][0m - global step 1480, epoch: 12, loss: 0.00035, speed: 0.80 step/s[0m
[32m[2022-10-20 22:21:19,940] [ INFO][0m - global step 1490, epoch: 12, loss: 0.00035, speed: 0.84 step/s[0m
[32m[2022-10-20 22:21:32,095] [ INFO][0m - global step 1500, epoch: 12, loss: 0.00034, speed: 0.82 step/s[0m
[32m[2022-10-20 22:21:33,476] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_1500/tokenizer_config.json[0m
[32m[2022-10-20 22:21:33,476] [ INFO][0m - Special tokens file saved in ./checkpoint/model_1500/special_tokens_map.json[0m
[32m[2022-10-20 22:21:49,900] [ INFO][0m - Evaluation precision: 0.90782, recall: 0.87367, F1: 0.89042[0m
[32m[2022-10-20 22:22:01,864] [ INFO][0m - global step 1510, epoch: 12, loss: 0.00034, speed: 0.84 step/s[0m
[32m[2022-10-20 22:22:14,140] [ INFO][0m - global step 1520, epoch: 12, loss: 0.00034, speed: 0.81 step/s[0m
[32m[2022-10-20 22:22:26,252] [ INFO][0m - global step 1530, epoch: 12, loss: 0.00034, speed: 0.83 step/s[0m
[32m[2022-10-20 22:22:38,431] [ INFO][0m - global step 1540, epoch: 12, loss: 0.00034, speed: 0.82 step/s[0m
[32m[2022-10-20 22:22:50,853] [ INFO][0m - global step 1550, epoch: 12, loss: 0.00034, speed: 0.81 step/s[0m
[32m[2022-10-20 22:22:52,330] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_1550/tokenizer_config.json[0m
[32m[2022-10-20 22:22:52,330] [ INFO][0m - Special tokens file saved in ./checkpoint/model_1550/special_tokens_map.json[0m
[32m[2022-10-20 22:23:09,234] [ INFO][0m - Evaluation precision: 0.88337, recall: 0.83992, F1: 0.86110[0m
[32m[2022-10-20 22:23:21,472] [ INFO][0m - global step 1560, epoch: 12, loss: 0.00034, speed: 0.82 step/s[0m
[32m[2022-10-20 22:23:33,754] [ INFO][0m - global step 1570, epoch: 12, loss: 0.00034, speed: 0.81 step/s[0m
[32m[2022-10-20 22:23:46,051] [ INFO][0m - global step 1580, epoch: 12, loss: 0.00034, speed: 0.81 step/s[0m
[32m[2022-10-20 22:23:58,128] [ INFO][0m - global step 1590, epoch: 12, loss: 0.00034, speed: 0.83 step/s[0m
[32m[2022-10-20 22:24:10,503] [ INFO][0m - global step 1600, epoch: 12, loss: 0.00033, speed: 0.81 step/s[0m
[32m[2022-10-20 22:24:11,757] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_1600/tokenizer_config.json[0m
[32m[2022-10-20 22:24:11,757] [ INFO][0m - Special tokens file saved in ./checkpoint/model_1600/special_tokens_map.json[0m
[32m[2022-10-20 22:24:27,622] [ INFO][0m - Evaluation precision: 0.88597, recall: 0.84667, F1: 0.86588[0m
[32m[2022-10-20 22:24:38,850] [ INFO][0m - global step 1610, epoch: 13, loss: 0.00033, speed: 0.89 step/s[0m
[32m[2022-10-20 22:24:50,925] [ INFO][0m - global step 1620, epoch: 13, loss: 0.00033, speed: 0.83 step/s[0m
[32m[2022-10-20 22:25:03,252] [ INFO][0m - global step 1630, epoch: 13, loss: 0.00033, speed: 0.81 step/s[0m
[32m[2022-10-20 22:25:15,089] [ INFO][0m - global step 1640, epoch: 13, loss: 0.00033, speed: 0.84 step/s[0m
[32m[2022-10-20 22:25:26,923] [ INFO][0m - global step 1650, epoch: 13, loss: 0.00033, speed: 0.85 step/s[0m
[32m[2022-10-20 22:25:28,478] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_1650/tokenizer_config.json[0m
[32m[2022-10-20 22:25:28,479] [ INFO][0m - Special tokens file saved in ./checkpoint/model_1650/special_tokens_map.json[0m
[32m[2022-10-20 22:25:44,957] [ INFO][0m - Evaluation precision: 0.87817, recall: 0.86885, F1: 0.87349[0m
[32m[2022-10-20 22:25:57,413] [ INFO][0m - global step 1660, epoch: 13, loss: 0.00033, speed: 0.80 step/s[0m
[32m[2022-10-20 22:26:09,977] [ INFO][0m - global step 1670, epoch: 13, loss: 0.00033, speed: 0.80 step/s[0m
[32m[2022-10-20 22:26:21,668] [ INFO][0m - global step 1680, epoch: 13, loss: 0.00032, speed: 0.86 step/s[0m
[32m[2022-10-20 22:26:33,732] [ INFO][0m - global step 1690, epoch: 13, loss: 0.00032, speed: 0.83 step/s[0m
[32m[2022-10-20 22:26:45,854] [ INFO][0m - global step 1700, epoch: 13, loss: 0.00032, speed: 0.82 step/s[0m
[32m[2022-10-20 22:26:47,129] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_1700/tokenizer_config.json[0m
[32m[2022-10-20 22:26:47,130] [ INFO][0m - Special tokens file saved in ./checkpoint/model_1700/special_tokens_map.json[0m
[32m[2022-10-20 22:27:03,520] [ INFO][0m - Evaluation precision: 0.86935, recall: 0.83414, F1: 0.85138[0m
[32m[2022-10-20 22:27:16,594] [ INFO][0m - global step 1710, epoch: 13, loss: 0.00032, speed: 0.76 step/s[0m
[32m[2022-10-20 22:27:28,840] [ INFO][0m - global step 1720, epoch: 13, loss: 0.00032, speed: 0.82 step/s[0m
[32m[2022-10-20 22:27:40,916] [ INFO][0m - global step 1730, epoch: 13, loss: 0.00032, speed: 0.83 step/s[0m
[32m[2022-10-20 22:27:52,528] [ INFO][0m - global step 1740, epoch: 13, loss: 0.00032, speed: 0.86 step/s[0m
[32m[2022-10-20 22:28:04,661] [ INFO][0m - global step 1750, epoch: 14, loss: 0.00032, speed: 0.82 step/s[0m
[32m[2022-10-20 22:28:05,888] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_1750/tokenizer_config.json[0m
[32m[2022-10-20 22:28:05,888] [ INFO][0m - Special tokens file saved in ./checkpoint/model_1750/special_tokens_map.json[0m
[32m[2022-10-20 22:28:21,818] [ INFO][0m - Evaluation precision: 0.89010, recall: 0.86692, F1: 0.87836[0m
[32m[2022-10-20 22:28:33,883] [ INFO][0m - global step 1760, epoch: 14, loss: 0.00032, speed: 0.83 step/s[0m
[32m[2022-10-20 22:28:45,993] [ INFO][0m - global step 1770, epoch: 14, loss: 0.00031, speed: 0.83 step/s[0m
[32m[2022-10-20 22:28:57,969] [ INFO][0m - global step 1780, epoch: 14, loss: 0.00031, speed: 0.84 step/s[0m
[32m[2022-10-20 22:29:10,313] [ INFO][0m - global step 1790, epoch: 14, loss: 0.00031, speed: 0.81 step/s[0m
[32m[2022-10-20 22:29:21,988] [ INFO][0m - global step 1800, epoch: 14, loss: 0.00031, speed: 0.86 step/s[0m
[32m[2022-10-20 22:29:23,201] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_1800/tokenizer_config.json[0m
[32m[2022-10-20 22:29:23,202] [ INFO][0m - Special tokens file saved in ./checkpoint/model_1800/special_tokens_map.json[0m
[32m[2022-10-20 22:29:40,163] [ INFO][0m - Evaluation precision: 0.88956, recall: 0.85439, F1: 0.87162[0m
[32m[2022-10-20 22:29:52,384] [ INFO][0m - global step 1810, epoch: 14, loss: 0.00031, speed: 0.82 step/s[0m
[32m[2022-10-20 22:30:04,547] [ INFO][0m - global step 1820, epoch: 14, loss: 0.00031, speed: 0.82 step/s[0m
[32m[2022-10-20 22:30:16,628] [ INFO][0m - global step 1830, epoch: 14, loss: 0.00030, speed: 0.83 step/s[0m
[32m[2022-10-20 22:30:28,812] [ INFO][0m - global step 1840, epoch: 14, loss: 0.00030, speed: 0.82 step/s[0m
[32m[2022-10-20 22:30:41,056] [ INFO][0m - global step 1850, epoch: 14, loss: 0.00030, speed: 0.82 step/s[0m
[32m[2022-10-20 22:30:42,272] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_1850/tokenizer_config.json[0m
[32m[2022-10-20 22:30:42,273] [ INFO][0m - Special tokens file saved in ./checkpoint/model_1850/special_tokens_map.json[0m
[32m[2022-10-20 22:30:57,475] [ INFO][0m - Evaluation precision: 0.89189, recall: 0.85921, F1: 0.87525[0m
[32m[2022-10-20 22:31:09,726] [ INFO][0m - global step 1860, epoch: 14, loss: 0.00030, speed: 0.82 step/s[0m
[32m[2022-10-20 22:31:21,554] [ INFO][0m - global step 1870, epoch: 14, loss: 0.00030, speed: 0.85 step/s[0m
[32m[2022-10-20 22:31:32,499] [ INFO][0m - global step 1880, epoch: 15, loss: 0.00030, speed: 0.91 step/s[0m
[32m[2022-10-20 22:31:44,957] [ INFO][0m - global step 1890, epoch: 15, loss: 0.00030, speed: 0.80 step/s[0m
[32m[2022-10-20 22:31:57,166] [ INFO][0m - global step 1900, epoch: 15, loss: 0.00030, speed: 0.82 step/s[0m
[32m[2022-10-20 22:31:58,382] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_1900/tokenizer_config.json[0m
[32m[2022-10-20 22:31:58,383] [ INFO][0m - Special tokens file saved in ./checkpoint/model_1900/special_tokens_map.json[0m
[32m[2022-10-20 22:32:15,448] [ INFO][0m - Evaluation precision: 0.89521, recall: 0.86500, F1: 0.87984[0m
[32m[2022-10-20 22:32:28,830] [ INFO][0m - global step 1910, epoch: 15, loss: 0.00029, speed: 0.75 step/s[0m
[32m[2022-10-20 22:32:40,834] [ INFO][0m - global step 1920, epoch: 15, loss: 0.00029, speed: 0.83 step/s[0m
[32m[2022-10-20 22:32:52,732] [ INFO][0m - global step 1930, epoch: 15, loss: 0.00029, speed: 0.84 step/s[0m
[32m[2022-10-20 22:33:04,746] [ INFO][0m - global step 1940, epoch: 15, loss: 0.00029, speed: 0.83 step/s[0m
[32m[2022-10-20 22:33:16,902] [ INFO][0m - global step 1950, epoch: 15, loss: 0.00029, speed: 0.82 step/s[0m
[32m[2022-10-20 22:33:18,131] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_1950/tokenizer_config.json[0m
[32m[2022-10-20 22:33:18,132] [ INFO][0m - Special tokens file saved in ./checkpoint/model_1950/special_tokens_map.json[0m
[32m[2022-10-20 22:33:34,806] [ INFO][0m - Evaluation precision: 0.88609, recall: 0.84764, F1: 0.86644[0m
[32m[2022-10-20 22:33:46,997] [ INFO][0m - global step 1960, epoch: 15, loss: 0.00029, speed: 0.82 step/s[0m
[32m[2022-10-20 22:33:59,397] [ INFO][0m - global step 1970, epoch: 15, loss: 0.00029, speed: 0.81 step/s[0m
[32m[2022-10-20 22:34:11,510] [ INFO][0m - global step 1980, epoch: 15, loss: 0.00029, speed: 0.83 step/s[0m
[32m[2022-10-20 22:34:23,226] [ INFO][0m - global step 1990, epoch: 15, loss: 0.00028, speed: 0.85 step/s[0m
[32m[2022-10-20 22:34:34,976] [ INFO][0m - global step 2000, epoch: 15, loss: 0.00028, speed: 0.85 step/s[0m
[32m[2022-10-20 22:34:36,219] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_2000/tokenizer_config.json[0m
[32m[2022-10-20 22:34:36,220] [ INFO][0m - Special tokens file saved in ./checkpoint/model_2000/special_tokens_map.json[0m
[32m[2022-10-20 22:34:52,486] [ INFO][0m - Evaluation precision: 0.89311, recall: 0.86210, F1: 0.87733[0m
[32m[2022-10-20 22:35:02,950] [ INFO][0m - global step 2010, epoch: 15, loss: 0.00028, speed: 0.96 step/s[0m
[32m[2022-10-20 22:35:15,876] [ INFO][0m - global step 2020, epoch: 16, loss: 0.00028, speed: 0.77 step/s[0m
[32m[2022-10-20 22:35:28,011] [ INFO][0m - global step 2030, epoch: 16, loss: 0.00028, speed: 0.82 step/s[0m
[32m[2022-10-20 22:35:40,220] [ INFO][0m - global step 2040, epoch: 16, loss: 0.00028, speed: 0.82 step/s[0m
[32m[2022-10-20 22:35:52,033] [ INFO][0m - global step 2050, epoch: 16, loss: 0.00028, speed: 0.85 step/s[0m
[32m[2022-10-20 22:35:53,258] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_2050/tokenizer_config.json[0m
[32m[2022-10-20 22:35:53,258] [ INFO][0m - Special tokens file saved in ./checkpoint/model_2050/special_tokens_map.json[0m
[32m[2022-10-20 22:36:09,582] [ INFO][0m - Evaluation precision: 0.88312, recall: 0.85246, F1: 0.86752[0m
[32m[2022-10-20 22:36:21,695] [ INFO][0m - global step 2060, epoch: 16, loss: 0.00028, speed: 0.83 step/s[0m
[32m[2022-10-20 22:36:33,879] [ INFO][0m - global step 2070, epoch: 16, loss: 0.00027, speed: 0.82 step/s[0m
[32m[2022-10-20 22:36:45,899] [ INFO][0m - global step 2080, epoch: 16, loss: 0.00027, speed: 0.83 step/s[0m
[32m[2022-10-20 22:36:58,150] [ INFO][0m - global step 2090, epoch: 16, loss: 0.00027, speed: 0.82 step/s[0m
[32m[2022-10-20 22:37:10,613] [ INFO][0m - global step 2100, epoch: 16, loss: 0.00027, speed: 0.80 step/s[0m
[32m[2022-10-20 22:37:11,863] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_2100/tokenizer_config.json[0m
[32m[2022-10-20 22:37:11,863] [ INFO][0m - Special tokens file saved in ./checkpoint/model_2100/special_tokens_map.json[0m
[32m[2022-10-20 22:37:28,199] [ INFO][0m - Evaluation precision: 0.90123, recall: 0.84474, F1: 0.87208[0m
[32m[2022-10-20 22:37:40,505] [ INFO][0m - global step 2110, epoch: 16, loss: 0.00027, speed: 0.81 step/s[0m
[32m[2022-10-20 22:37:52,178] [ INFO][0m - global step 2120, epoch: 16, loss: 0.00027, speed: 0.86 step/s[0m
[32m[2022-10-20 22:38:04,484] [ INFO][0m - global step 2130, epoch: 16, loss: 0.00027, speed: 0.81 step/s[0m
[32m[2022-10-20 22:38:16,613] [ INFO][0m - global step 2140, epoch: 16, loss: 0.00027, speed: 0.82 step/s[0m
[32m[2022-10-20 22:38:28,037] [ INFO][0m - global step 2150, epoch: 17, loss: 0.00026, speed: 0.88 step/s[0m
[32m[2022-10-20 22:38:29,264] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_2150/tokenizer_config.json[0m
[32m[2022-10-20 22:38:29,264] [ INFO][0m - Special tokens file saved in ./checkpoint/model_2150/special_tokens_map.json[0m
[32m[2022-10-20 22:38:46,339] [ INFO][0m - Evaluation precision: 0.89990, recall: 0.85824, F1: 0.87858[0m
[32m[2022-10-20 22:38:58,648] [ INFO][0m - global step 2160, epoch: 17, loss: 0.00026, speed: 0.81 step/s[0m
[32m[2022-10-20 22:39:10,945] [ INFO][0m - global step 2170, epoch: 17, loss: 0.00026, speed: 0.81 step/s[0m
[32m[2022-10-20 22:39:22,923] [ INFO][0m - global step 2180, epoch: 17, loss: 0.00026, speed: 0.83 step/s[0m
[32m[2022-10-20 22:39:34,907] [ INFO][0m - global step 2190, epoch: 17, loss: 0.00026, speed: 0.83 step/s[0m
[32m[2022-10-20 22:39:47,522] [ INFO][0m - global step 2200, epoch: 17, loss: 0.00026, speed: 0.79 step/s[0m
[32m[2022-10-20 22:39:48,727] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_2200/tokenizer_config.json[0m
[32m[2022-10-20 22:39:48,727] [ INFO][0m - Special tokens file saved in ./checkpoint/model_2200/special_tokens_map.json[0m
[32m[2022-10-20 22:40:06,299] [ INFO][0m - Evaluation precision: 0.89076, recall: 0.86500, F1: 0.87769[0m
[32m[2022-10-20 22:40:18,623] [ INFO][0m - global step 2210, epoch: 17, loss: 0.00026, speed: 0.81 step/s[0m
[32m[2022-10-20 22:40:30,427] [ INFO][0m - global step 2220, epoch: 17, loss: 0.00026, speed: 0.85 step/s[0m
[32m[2022-10-20 22:40:42,625] [ INFO][0m - global step 2230, epoch: 17, loss: 0.00026, speed: 0.82 step/s[0m
[32m[2022-10-20 22:40:54,544] [ INFO][0m - global step 2240, epoch: 17, loss: 0.00026, speed: 0.84 step/s[0m
[32m[2022-10-20 22:41:07,015] [ INFO][0m - global step 2250, epoch: 17, loss: 0.00026, speed: 0.80 step/s[0m
[32m[2022-10-20 22:41:08,346] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_2250/tokenizer_config.json[0m
[32m[2022-10-20 22:41:08,346] [ INFO][0m - Special tokens file saved in ./checkpoint/model_2250/special_tokens_map.json[0m
[32m[2022-10-20 22:41:24,625] [ INFO][0m - Evaluation precision: 0.88786, recall: 0.83221, F1: 0.85913[0m
[32m[2022-10-20 22:41:37,054] [ INFO][0m - global step 2260, epoch: 17, loss: 0.00026, speed: 0.80 step/s[0m
[32m[2022-10-20 22:41:49,518] [ INFO][0m - global step 2270, epoch: 17, loss: 0.00026, speed: 0.80 step/s[0m
[32m[2022-10-20 22:42:01,357] [ INFO][0m - global step 2280, epoch: 18, loss: 0.00026, speed: 0.84 step/s[0m
[32m[2022-10-20 22:42:13,812] [ INFO][0m - global step 2290, epoch: 18, loss: 0.00026, speed: 0.80 step/s[0m
[32m[2022-10-20 22:42:26,240] [ INFO][0m - global step 2300, epoch: 18, loss: 0.00026, speed: 0.80 step/s[0m
[32m[2022-10-20 22:42:27,456] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_2300/tokenizer_config.json[0m
[32m[2022-10-20 22:42:27,457] [ INFO][0m - Special tokens file saved in ./checkpoint/model_2300/special_tokens_map.json[0m
[32m[2022-10-20 22:42:45,058] [ INFO][0m - Evaluation precision: 0.88732, recall: 0.85053, F1: 0.86854[0m
[32m[2022-10-20 22:42:57,009] [ INFO][0m - global step 2310, epoch: 18, loss: 0.00026, speed: 0.84 step/s[0m
[32m[2022-10-20 22:43:09,380] [ INFO][0m - global step 2320, epoch: 18, loss: 0.00026, speed: 0.81 step/s[0m
[32m[2022-10-20 22:43:21,508] [ INFO][0m - global step 2330, epoch: 18, loss: 0.00026, speed: 0.82 step/s[0m
[32m[2022-10-20 22:43:33,886] [ INFO][0m - global step 2340, epoch: 18, loss: 0.00026, speed: 0.81 step/s[0m
[32m[2022-10-20 22:43:46,374] [ INFO][0m - global step 2350, epoch: 18, loss: 0.00026, speed: 0.80 step/s[0m
[32m[2022-10-20 22:43:47,719] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_2350/tokenizer_config.json[0m
[32m[2022-10-20 22:43:47,720] [ INFO][0m - Special tokens file saved in ./checkpoint/model_2350/special_tokens_map.json[0m
[32m[2022-10-20 22:44:04,405] [ INFO][0m - Evaluation precision: 0.88100, recall: 0.84957, F1: 0.86500[0m
[32m[2022-10-20 22:44:16,557] [ INFO][0m - global step 2360, epoch: 18, loss: 0.00026, speed: 0.82 step/s[0m
[32m[2022-10-20 22:44:29,025] [ INFO][0m - global step 2370, epoch: 18, loss: 0.00026, speed: 0.80 step/s[0m
[32m[2022-10-20 22:44:41,328] [ INFO][0m - global step 2380, epoch: 18, loss: 0.00026, speed: 0.81 step/s[0m
[32m[2022-10-20 22:44:53,314] [ INFO][0m - global step 2390, epoch: 18, loss: 0.00026, speed: 0.83 step/s[0m
[32m[2022-10-20 22:45:05,686] [ INFO][0m - global step 2400, epoch: 18, loss: 0.00026, speed: 0.81 step/s[0m
[32m[2022-10-20 22:45:07,004] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_2400/tokenizer_config.json[0m
[32m[2022-10-20 22:45:07,005] [ INFO][0m - Special tokens file saved in ./checkpoint/model_2400/special_tokens_map.json[0m
[32m[2022-10-20 22:45:24,074] [ INFO][0m - Evaluation precision: 0.89122, recall: 0.86114, F1: 0.87592[0m
[32m[2022-10-20 22:45:35,529] [ INFO][0m - global step 2410, epoch: 18, loss: 0.00026, speed: 0.87 step/s[0m
[32m[2022-10-20 22:45:47,387] [ INFO][0m - global step 2420, epoch: 19, loss: 0.00026, speed: 0.84 step/s[0m
[32m[2022-10-20 22:45:59,962] [ INFO][0m - global step 2430, epoch: 19, loss: 0.00026, speed: 0.80 step/s[0m
[32m[2022-10-20 22:46:12,592] [ INFO][0m - global step 2440, epoch: 19, loss: 0.00026, speed: 0.79 step/s[0m
[32m[2022-10-20 22:46:24,411] [ INFO][0m - global step 2450, epoch: 19, loss: 0.00026, speed: 0.85 step/s[0m
[32m[2022-10-20 22:46:25,653] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_2450/tokenizer_config.json[0m
[32m[2022-10-20 22:46:25,653] [ INFO][0m - Special tokens file saved in ./checkpoint/model_2450/special_tokens_map.json[0m
[32m[2022-10-20 22:46:42,121] [ INFO][0m - Evaluation precision: 0.88569, recall: 0.85921, F1: 0.87225[0m
[32m[2022-10-20 22:46:54,557] [ INFO][0m - global step 2460, epoch: 19, loss: 0.00026, speed: 0.80 step/s[0m
[32m[2022-10-20 22:47:06,761] [ INFO][0m - global step 2470, epoch: 19, loss: 0.00026, speed: 0.82 step/s[0m
[32m[2022-10-20 22:47:19,231] [ INFO][0m - global step 2480, epoch: 19, loss: 0.00026, speed: 0.80 step/s[0m
[32m[2022-10-20 22:47:31,271] [ INFO][0m - global step 2490, epoch: 19, loss: 0.00026, speed: 0.83 step/s[0m
[32m[2022-10-20 22:47:43,382] [ INFO][0m - global step 2500, epoch: 19, loss: 0.00026, speed: 0.83 step/s[0m
[32m[2022-10-20 22:47:44,596] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_2500/tokenizer_config.json[0m
[32m[2022-10-20 22:47:44,596] [ INFO][0m - Special tokens file saved in ./checkpoint/model_2500/special_tokens_map.json[0m
[32m[2022-10-20 22:48:02,284] [ INFO][0m - Evaluation precision: 0.87132, recall: 0.85535, F1: 0.86326[0m
[32m[2022-10-20 22:48:14,664] [ INFO][0m - global step 2510, epoch: 19, loss: 0.00026, speed: 0.81 step/s[0m
[32m[2022-10-20 22:48:26,749] [ INFO][0m - global step 2520, epoch: 19, loss: 0.00026, speed: 0.83 step/s[0m
[32m[2022-10-20 22:48:39,293] [ INFO][0m - global step 2530, epoch: 19, loss: 0.00026, speed: 0.80 step/s[0m
[32m[2022-10-20 22:48:51,329] [ INFO][0m - global step 2540, epoch: 19, loss: 0.00025, speed: 0.83 step/s[0m
[32m[2022-10-20 22:49:01,877] [ INFO][0m - global step 2550, epoch: 20, loss: 0.00025, speed: 0.95 step/s[0m
[32m[2022-10-20 22:49:03,441] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_2550/tokenizer_config.json[0m
[32m[2022-10-20 22:49:03,441] [ INFO][0m - Special tokens file saved in ./checkpoint/model_2550/special_tokens_map.json[0m
[32m[2022-10-20 22:49:20,552] [ INFO][0m - Evaluation precision: 0.88922, recall: 0.85921, F1: 0.87396[0m
[32m[2022-10-20 22:49:33,100] [ INFO][0m - global step 2560, epoch: 20, loss: 0.00025, speed: 0.80 step/s[0m
[32m[2022-10-20 22:49:44,738] [ INFO][0m - global step 2570, epoch: 20, loss: 0.00025, speed: 0.86 step/s[0m
[32m[2022-10-20 22:49:55,939] [ INFO][0m - global step 2580, epoch: 20, loss: 0.00025, speed: 0.89 step/s[0m
[32m[2022-10-20 22:50:07,832] [ INFO][0m - global step 2590, epoch: 20, loss: 0.00025, speed: 0.84 step/s[0m
[32m[2022-10-20 22:50:19,708] [ INFO][0m - global step 2600, epoch: 20, loss: 0.00025, speed: 0.84 step/s[0m
[32m[2022-10-20 22:50:21,176] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_2600/tokenizer_config.json[0m
[32m[2022-10-20 22:50:21,177] [ INFO][0m - Special tokens file saved in ./checkpoint/model_2600/special_tokens_map.json[0m
[32m[2022-10-20 22:50:38,368] [ INFO][0m - Evaluation precision: 0.89395, recall: 0.86982, F1: 0.88172[0m
[32m[2022-10-20 22:50:50,843] [ INFO][0m - global step 2610, epoch: 20, loss: 0.00025, speed: 0.80 step/s[0m
[32m[2022-10-20 22:51:04,090] [ INFO][0m - global step 2620, epoch: 20, loss: 0.00025, speed: 0.75 step/s[0m
[32m[2022-10-20 22:51:16,512] [ INFO][0m - global step 2630, epoch: 20, loss: 0.00025, speed: 0.81 step/s[0m
[32m[2022-10-20 22:51:28,512] [ INFO][0m - global step 2640, epoch: 20, loss: 0.00025, speed: 0.83 step/s[0m
[32m[2022-10-20 22:51:40,770] [ INFO][0m - global step 2650, epoch: 20, loss: 0.00025, speed: 0.82 step/s[0m
[32m[2022-10-20 22:51:41,977] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_2650/tokenizer_config.json[0m
[32m[2022-10-20 22:51:41,978] [ INFO][0m - Special tokens file saved in ./checkpoint/model_2650/special_tokens_map.json[0m
[32m[2022-10-20 22:51:59,960] [ INFO][0m - Evaluation precision: 0.89163, recall: 0.87271, F1: 0.88207[0m
[32m[2022-10-20 22:52:12,226] [ INFO][0m - global step 2660, epoch: 20, loss: 0.00024, speed: 0.82 step/s[0m
[32m[2022-10-20 22:52:24,316] [ INFO][0m - global step 2670, epoch: 20, loss: 0.00024, speed: 0.83 step/s[0m
[32m[2022-10-20 22:52:34,724] [ INFO][0m - global step 2680, epoch: 20, loss: 0.00024, speed: 0.96 step/s[0m
[32m[2022-10-20 22:52:48,075] [ INFO][0m - global step 2690, epoch: 21, loss: 0.00024, speed: 0.75 step/s[0m
[32m[2022-10-20 22:53:00,661] [ INFO][0m - global step 2700, epoch: 21, loss: 0.00024, speed: 0.79 step/s[0m
[32m[2022-10-20 22:53:01,844] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_2700/tokenizer_config.json[0m
[32m[2022-10-20 22:53:01,844] [ INFO][0m - Special tokens file saved in ./checkpoint/model_2700/special_tokens_map.json[0m
[32m[2022-10-20 22:53:19,543] [ INFO][0m - Evaluation precision: 0.89243, recall: 0.86403, F1: 0.87800[0m
[32m[2022-10-20 22:53:31,644] [ INFO][0m - global step 2710, epoch: 21, loss: 0.00024, speed: 0.83 step/s[0m
[32m[2022-10-20 22:53:43,600] [ INFO][0m - global step 2720, epoch: 21, loss: 0.00024, speed: 0.84 step/s[0m
[32m[2022-10-20 22:53:55,880] [ INFO][0m - global step 2730, epoch: 21, loss: 0.00024, speed: 0.81 step/s[0m
[32m[2022-10-20 22:54:08,212] [ INFO][0m - global step 2740, epoch: 21, loss: 0.00024, speed: 0.81 step/s[0m
[32m[2022-10-20 22:54:20,216] [ INFO][0m - global step 2750, epoch: 21, loss: 0.00024, speed: 0.83 step/s[0m
[32m[2022-10-20 22:54:21,488] [ INFO][0m - tokenizer config file saved in ./checkpoint/model_2750/tokenizer_config.json[0m
[32m[2022-10-20 22:54:21,489] [ INFO][0m - Special tokens file saved in ./checkpoint/model_2750/special_tokens_map.json[0m
[32m[2022-10-20 22:54:38,048] [ INFO][0m - Evaluation precision: 0.89000, recall: 0.85824, F1: 0.87383[0m
[32m[2022-10-20 22:54:50,184] [ INFO][0m - global step 2760, epoch: 21, loss: 0.00024, speed: 0.82 step/s[0m
[32m[2022-10-20 22:55:02,255] [ INFO][0m - global step 2770, epoch: 21, loss: 0.00024, speed: 0.83 step/s[0m
[32m[2022-10-20 22:55:14,814] [ INFO][0m - global step 2780, epoch: 21, loss: 0.00024, speed: 0.80 step/s[0m
^C
Traceback (most recent call last):
File "finetune.py", line 173, in <module>
do_train()
File "finetune.py", line 106, in do_train
start_ids = paddle.cast(start_ids, 'float32')
File "/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/tensor/manipulation.py", line 67, in cast
return _C_ops.cast(x, dtype)
KeyboardInterrupt
[0m
训练日志
[2022-10-20 16:57:57,110] [ INFO] - global step 570, epoch: 14, loss: 0.00189, speed: 0.76 step/s
[2022-10-20 16:58:09,910] [ INFO] - global step 580, epoch: 14, loss: 0.00188, speed: 0.78 step/s
[2022-10-20 16:58:19,367] [ INFO] - Evaluation precision: 0.84211, recall: 0.77859, F1: 0.80910
[2022-10-20 16:58:19,367] [ INFO] - best F1 performence has been updated: 0.80856 --> 0.80910
3.模型评估
!python evaluate.py \
--model_path ./checkpoint/model_best \
--test_path data_save/dev.txt \
--batch_size 40 \
--max_seq_len 512
[32m[2022-10-20 22:56:01,067] [ INFO][0m - We are using <class 'paddlenlp.transformers.ernie.tokenizer.ErnieTokenizer'> to load './checkpoint/model_best'.[0m
W1020 22:56:01.100837 19956 gpu_resources.cc:61] Please NOTE: device: 0, GPU Compute Capability: 7.0, Driver API Version: 11.2, Runtime API Version: 11.2
W1020 22:56:01.104740 19956 gpu_resources.cc:91] device: 0, cuDNN Version: 8.2.
[32m[2022-10-20 22:56:21,524] [ INFO][0m - -----------------------------[0m
[32m[2022-10-20 22:56:21,524] [ INFO][0m - Class Name: all_classes[0m
[32m[2022-10-20 22:56:21,524] [ INFO][0m - Evaluation Precision: 0.91089 | Recall: 0.88717 | F1: 0.89888[0m
[0m
!python evaluate.py \
--model_path ./checkpoint/model_2750 \
--test_path data_save/dev.txt \
--batch_size 40 \
--max_seq_len 512
[32m[2022-10-20 22:56:34,946] [ INFO][0m - We are using <class 'paddlenlp.transformers.ernie.tokenizer.ErnieTokenizer'> to load './checkpoint/model_2750'.[0m
W1020 22:56:34.978999 20083 gpu_resources.cc:61] Please NOTE: device: 0, GPU Compute Capability: 7.0, Driver API Version: 11.2, Runtime API Version: 11.2
W1020 22:56:34.989790 20083 gpu_resources.cc:91] device: 0, cuDNN Version: 8.2.
[32m[2022-10-20 22:56:56,038] [ INFO][0m - -----------------------------[0m
[32m[2022-10-20 22:56:56,038] [ INFO][0m - Class Name: all_classes[0m
[32m[2022-10-20 22:56:56,038] [ INFO][0m - Evaluation Precision: 0.89000 | Recall: 0.85824 | F1: 0.87383[0m
[0m
四、结果预测
读取数据并送入模型预测
# 删除jupyter隐藏文件
!rm 测试集_选手/.ipynb* -rf
!rm 测试集_选手/案件要素/.ipynb* -rf
!ls 测试集_选手 -la
总用量 173
drwxr-xr-x 1 aistudio aistudio 4096 10月 20 18:21 .
drwxr-xr-x 1 aistudio aistudio 4096 10月 20 18:47 ..
-rw-r--r-- 1 aistudio aistudio 66777 9月 14 18:52 ner
drwxr-xr-x 1 aistudio aistudio 4096 10月 20 18:47 案件要素
-rw-r--r-- 1 aistudio aistudio 97580 9月 14 18:52 刑档
!ls 测试集_选手/案件要素 -la
总用量 345
drwxr-xr-x 1 aistudio aistudio 4096 10月 20 18:47 .
drwxr-xr-x 1 aistudio aistudio 4096 10月 20 18:21 ..
-rw-r--r-- 1 aistudio aistudio 52434 9月 14 18:52 被害人被后车撞击
-rw-r--r-- 1 aistudio aistudio 35406 9月 14 18:52 被害人闯红灯
-rw-r--r-- 1 aistudio aistudio 89642 9月 14 18:52 被害人为本车人员
-rw-r--r-- 1 aistudio aistudio 31478 9月 14 18:52 交通肇事后逃逸
-rw-r--r-- 1 aistudio aistudio 27156 9月 14 18:52 全部责任
-rw-r--r-- 1 aistudio aistudio 30030 9月 14 18:52 肇事车辆超速行驶
-rw-r--r-- 1 aistudio aistudio 34194 9月 14 18:52 肇事车辆逆行
-rw-r--r-- 1 aistudio aistudio 41809 9月 14 18:52 中型客车交通肇事
#coding:utf-8
import json
import os
import argparse
from paddlenlp import Taskflow
def data_read(path):
data_dir = {}
for file_name in os.listdir(path):
if file_name != '案件要素':
data_dir[file_name] = []
with open(os.path.join(path, file_name), encoding='utf8') as f:
for line in f:
line_js = json.loads(line)
data_dir[file_name].append(line_js)
else:
for case_plot_name in os.listdir(os.path.join(path, file_name)):
data_dir[case_plot_name] = []
with open(os.path.join(path, os.path.join(file_name, case_plot_name)), encoding='utf8') as f:
for line in f:
line_js = json.loads(line)
data_dir[case_plot_name].append(line_js)
return data_dir
def predict(input_dir, output_file):
data = data_read(input_dir)
print(len(data))
print(data.keys())
with open(output_file, 'w', encoding='utf8') as output:
for task_name,task_data_list in data.items():
print("task_name: ", task_name)
if task_name != 'ner':
tq_name = 'data'
if task_name == '刑档':
prompt = task_name + '[一档,二档,三档]'
else:
prompt = task_name + '[正,负]'
else:
tq_name = 'text'
prompt = ' '
ie = Taskflow(task='information_extraction', schema=prompt, task_path='checkpoint/model_best')
if prompt != ' ': # 分类任务
for line in task_data_list:
try:
if task_name == '刑档':
line_data=line[tq_name].strip('\n').strip().replace('[SEP]','',999)
else :
line_data=line[tq_name].strip('\n').strip()
result = ie(line_data)
print(line['id'], result)
p_lab = [[result[0][prompt][0]['text']]]
except:
p_lab = [[]]
output.write(json.dumps({'id':line['id'], 'data':line[tq_name], 'task':line['task'], 'label':p_lab}, ensure_ascii=False)+'\n')
else: # ner任务
prompt = ['犯罪嫌疑人情况','被害人','被害人类型','犯罪嫌疑人交通工具','犯罪嫌疑人交通工具情况','被害人交通工具情况','犯罪嫌疑人责任认定','被害人责任认定','事故发生地','被害人交通工具']
ie = Taskflow(task='information_extraction', schema=prompt, task_path='checkpoint/model_best')
for line in task_data_list:
try:
p_lab={}
line_data=line[tq_name].strip('\n').strip()
result = ie(line_data)
print(line['id'], result)
for entity_type in prompt:
p_lab[entity_type] = []
if entity_type not in result[0]:
p_lab[entity_type].append([])
continue
for samp in result[0][entity_type]:
p_lab[entity_type].append([samp['start'], samp['end']])
except:
continue
output.write(json.dumps({'id':line['id'], 'data':line[tq_name], 'task':line['task'], 'label':p_lab}, ensure_ascii=False)+'\n')
predict('测试集_选手/', 'result.txt')
10
dict_keys(['被害人闯红灯', '被害人被后车撞击', '肇事车辆逆行', '全部责任', '交通肇事后逃逸', '肇事车辆超速行驶', '被害人为本车人员', '中型客车交通肇事', 'ner', '刑档'])
task_name: 被害人闯红灯
[2022-10-20 22:57:23,568] [ INFO] - We are using <class 'paddlenlp.transformers.ernie.tokenizer.ErnieTokenizer'> to load 'checkpoint/model_best'.
W1020 22:57:23.603516 10330 gpu_resources.cc:61] Please NOTE: device: 0, GPU Compute Capability: 7.0, Driver API Version: 11.2, Runtime API Version: 11.2
W1020 22:57:23.607884 10330 gpu_resources.cc:91] device: 0, cuDNN Version: 8.2.
[2022-10-20 22:57:26,851] [ INFO] - Converting to the inference model cost a little time.
[2022-10-20 22:57:38,800] [ INFO] - The inference model save in the path:checkpoint/model_best/static/inference
346 [{'被害人闯红灯[正,负]': [{'text': '负', 'probability': 0.9988890118593297}]}]
347 [{'被害人闯红灯[正,负]': [{'text': '负', 'probability': 0.9990447474161783}]}]
348 [{'被害人闯红灯[正,负]': [{'text': '负', 'probability': 0.9993371117178071}]}]
349 [{'被害人闯红灯[正,负]': [{'text': '负', 'probability': 0.9970991993934651}]}]
350 [{'被害人闯红灯[正,负]': [{'text': '负', 'probability': 0.9962109602222213}]}]
351 [{'被害人闯红灯[正,负]': [{'text': '负', 'probability': 0.9992121736780746}]}]
352 [{'被害人闯红灯[正,负]': [{'text': '负', 'probability': 0.9992019827065945}]}]
353 [{'被害人闯红灯[正,负]': [{'text': '负', 'probability': 0.9992211778438147}]}]
354 [{'被害人闯红灯[正,负]': [{'text': '负', 'probability': 0.9993440867860421}]}]
355 [{'被害人闯红灯[正,负]': [{'text': '负', 'probability': 0.9944282947338792}]}]
356 [{'被害人闯红灯[正,负]': [{'text': '负', 'probability': 0.9396762208787663}]}]
357 [{'被害人闯红灯[正,负]': [{'text': '负', 'probability': 0.9992057930202733}]}]
358 [{'被害人闯红灯[正,负]': [{'text': '负', 'probability': 0.9961358146361352}]}]
359 [{'被害人闯红灯[正,负]': [{'text': '负', 'probability': 0.9993759728746028}]}]
360 [{'被害人闯红灯[正,负]': [{'text': '负', 'probability': 0.9993871024627303}]}]
361 [{'被害人闯红灯[正,负]': [{'text': '负', 'probability': 0.9992234052078146}]}]
362 [{'被害人闯红灯[正,负]': [{'text': '负', 'probability': 0.997425891971325}]}]
363 [{'被害人闯红灯[正,负]': [{'text': '负', 'probability': 0.9992955971728179}]}]
364 [{'被害人闯红灯[正,负]': [{'text': '负', 'probability': 0.9992490620108185}]}]
365 [{'被害人闯红灯[正,负]': [{'text': '负', 'probability': 0.9989217757867266}]}]
366 [{'被害人闯红灯[正,负]': [{'text': '负', 'probability': 0.9987431771089916}]}]
367 [{'被害人闯红灯[正,负]': [{'text': '负', 'probability': 0.9993785300123754}]}]
368 [{'被害人闯红灯[正,负]': [{'text': '负', 'probability': 0.9968313479254576}]}]
369 [{'被害人闯红灯[正,负]': [{'text': '负', 'probability': 0.9990783066765374}]}]
370 [{'被害人闯红灯[正,负]': [{'text': '负', 'probability': 0.9993089396946964}]}]
371 [{'被害人闯红灯[正,负]': [{'text': '负', 'probability': 0.9992038978986244}]}]
372 [{'被害人闯红灯[正,负]': [{'text': '负', 'probability': 0.407962882963254}]}]
task_name: 被害人被后车撞击
[2022-10-20 22:57:42,202] [ INFO] - We are using <class 'paddlenlp.transformers.ernie.tokenizer.ErnieTokenizer'> to load 'checkpoint/model_best'.
318 [{'被害人被后车撞击[正,负]': [{'text': '正', 'probability': 0.9862992484329496}]}]
319 [{'被害人被后车撞击[正,负]': [{'text': '正', 'probability': 0.9991880824015364}]}]
320 [{'被害人被后车撞击[正,负]': [{'text': '负', 'probability': 0.8806649750059137}]}]
321 [{'被害人被后车撞击[正,负]': [{'text': '正', 'probability': 0.9949399652849849}]}]
322 [{'被害人被后车撞击[正,负]': [{'text': '正', 'probability': 0.9971320338971736}]}]
323 [{'被害人被后车撞击[正,负]': [{'text': '负', 'probability': 0.9931023121595501}]}]
324 [{'被害人被后车撞击[正,负]': [{'text': '正', 'probability': 0.9286047166339202}]}]
325 [{'被害人被后车撞击[正,负]': [{'text': '正', 'probability': 0.9982464838398499}]}]
326 [{'被害人被后车撞击[正,负]': [{'text': '负', 'probability': 0.9982623421601033}]}]
327 [{'被害人被后车撞击[正,负]': [{'text': '负', 'probability': 0.9533525832274066}]}]
328 [{'被害人被后车撞击[正,负]': [{'text': '负', 'probability': 0.9957978680610537}]}]
329 [{'被害人被后车撞击[正,负]': [{'text': '正', 'probability': 0.9957680035004639}]}]
330 [{'被害人被后车撞击[正,负]': [{'text': '正', 'probability': 0.7759313274069655}]}]
331 [{'被害人被后车撞击[正,负]': [{'text': '正', 'probability': 0.9994054859579009}]}]
332 [{'被害人被后车撞击[正,负]': [{'text': '负', 'probability': 0.9986162042122793}]}]
333 [{'被害人被后车撞击[正,负]': [{'text': '负', 'probability': 0.9749774502600346}]}]
334 [{'被害人被后车撞击[正,负]': [{'text': '负', 'probability': 0.9954518604624241}]}]
335 [{'被害人被后车撞击[正,负]': [{'text': '负', 'probability': 0.9744195813710554}]}]
336 [{'被害人被后车撞击[正,负]': [{'text': '负', 'probability': 0.9538589937184199}]}]
337 [{'被害人被后车撞击[正,负]': [{'text': '负', 'probability': 0.993331657901269}]}]
338 [{'被害人被后车撞击[正,负]': [{'text': '正', 'probability': 0.9456753632422206}]}]
339 [{'被害人被后车撞击[正,负]': [{'text': '正', 'probability': 0.9254473061479302}]}]
340 [{'被害人被后车撞击[正,负]': [{'text': '负', 'probability': 0.9970967878058588}]}]
341 [{'被害人被后车撞击[正,负]': [{'text': '负', 'probability': 0.9302808006925574}]}]
342 [{'被害人被后车撞击[正,负]': [{'text': '负', 'probability': 0.9949222307353125}]}]
343 [{'被害人被后车撞击[正,负]': [{'text': '正', 'probability': 0.3903173544683227}]}]
344 [{'被害人被后车撞击[正,负]': [{'text': '正', 'probability': 0.9847168487149247}]}]
345 [{'被害人被后车撞击[正,负]': [{'text': '正', 'probability': 0.9979761276914587}]}]
task_name: 肇事车辆逆行
[2022-10-20 22:57:45,557] [ INFO] - We are using <class 'paddlenlp.transformers.ernie.tokenizer.ErnieTokenizer'> to load 'checkpoint/model_best'.
260 [{'肇事车辆逆行[正,负]': [{'text': '正', 'probability': 0.997946390802646}]}]
261 [{'肇事车辆逆行[正,负]': [{'text': '正', 'probability': 0.9854103358764021}]}]
262 [{'肇事车辆逆行[正,负]': [{'text': '正', 'probability': 0.9991487881053551}]}]
263 [{'肇事车辆逆行[正,负]': [{'text': '正', 'probability': 0.9301457101100752}]}]
264 [{'肇事车辆逆行[正,负]': [{'text': '负', 'probability': 0.8488694489554582}]}]
265 [{'肇事车辆逆行[正,负]': [{'text': '正', 'probability': 0.9991090429649319}]}]
266 [{}]
267 [{'肇事车辆逆行[正,负]': [{'text': '正', 'probability': 0.9884654806691273}]}]
268 [{'肇事车辆逆行[正,负]': [{'text': '负', 'probability': 0.9982471357190086}]}]
269 [{'肇事车辆逆行[正,负]': [{'text': '正', 'probability': 0.9990637694211202}]}]
270 [{'肇事车辆逆行[正,负]': [{'text': '负', 'probability': 0.9963652023720755}]}]
271 [{'肇事车辆逆行[正,负]': [{'text': '负', 'probability': 0.9989578832214505}]}]
272 [{'肇事车辆逆行[正,负]': [{'text': '负', 'probability': 0.9986492489993033}]}]
273 [{'肇事车辆逆行[正,负]': [{'text': '负', 'probability': 0.9919454318410388}]}]
274 [{'肇事车辆逆行[正,负]': [{'text': '负', 'probability': 0.9787970398079722}]}]
275 [{'肇事车辆逆行[正,负]': [{'text': '负', 'probability': 0.9992727572531734}]}]
276 [{'肇事车辆逆行[正,负]': [{'text': '负', 'probability': 0.9983147358696876}]}]
277 [{'肇事车辆逆行[正,负]': [{'text': '负', 'probability': 0.9990897206449034}]}]
278 [{'肇事车辆逆行[正,负]': [{'text': '负', 'probability': 0.9992498867792179}]}]
279 [{'肇事车辆逆行[正,负]': [{'text': '负', 'probability': 0.9985632450955073}]}]
280 [{'肇事车辆逆行[正,负]': [{'text': '负', 'probability': 0.9989133652515108}]}]
281 [{'肇事车辆逆行[正,负]': [{'text': '负', 'probability': 0.9992097830133666}]}]
282 [{'肇事车辆逆行[正,负]': [{'text': '负', 'probability': 0.9992502472049978}]}]
283 [{'肇事车辆逆行[正,负]': [{'text': '负', 'probability': 0.999144048827759}]}]
284 [{'肇事车辆逆行[正,负]': [{'text': '正', 'probability': 0.983831431122578}]}]
285 [{'肇事车辆逆行[正,负]': [{'text': '负', 'probability': 0.9985279544823946}]}]
286 [{'肇事车辆逆行[正,负]': [{'text': '负', 'probability': 0.999250005614968}]}]
task_name: 全部责任
[2022-10-20 22:57:48,046] [ INFO] - We are using <class 'paddlenlp.transformers.ernie.tokenizer.ErnieTokenizer'> to load 'checkpoint/model_best'.
213 [{'全部责任[正,负]': [{'text': '正', 'probability': 0.9892957536283902}]}]
214 [{'全部责任[正,负]': [{'text': '正', 'probability': 0.999192935095472}]}]
215 [{'全部责任[正,负]': [{'text': '正', 'probability': 0.9108441027183076}]}]
216 [{'全部责任[正,负]': [{'text': '正', 'probability': 0.998571280684903}]}]
217 [{'全部责任[正,负]': [{'text': '正', 'probability': 0.9690052991957572}]}]
218 [{'全部责任[正,负]': [{'text': '正', 'probability': 0.9991560571383395}]}]
219 [{'全部责任[正,负]': [{'text': '正', 'probability': 0.9994384144532376}]}]
220 [{'全部责任[正,负]': [{'text': '正', 'probability': 0.9974441879019906}]}]
221 [{'全部责任[正,负]': [{'text': '正', 'probability': 0.9996086740629018}]}]
222 [{'全部责任[正,负]': [{'text': '正', 'probability': 0.999143692062539}]}]
223 [{'全部责任[正,负]': [{'text': '正', 'probability': 0.9990799724420398}]}]
224 [{'全部责任[正,负]': [{'text': '负', 'probability': 0.29055310073668394}]}]
225 [{'全部责任[正,负]': [{'text': '正', 'probability': 0.9995343591318218}]}]
226 [{'全部责任[正,负]': [{'text': '正', 'probability': 0.9995352565927824}]}]
227 [{'全部责任[正,负]': [{'text': '负', 'probability': 0.9993240781721511}]}]
228 [{'全部责任[正,负]': [{'text': '负', 'probability': 0.993928208681556}]}]
229 [{'全部责任[正,负]': [{'text': '负', 'probability': 0.9988451761335959}]}]
230 [{'全部责任[正,负]': [{'text': '负', 'probability': 0.9977627358049546}]}]
231 [{'全部责任[正,负]': [{'text': '负', 'probability': 0.3760157664082442}]}]
232 [{'全部责任[正,负]': [{'text': '负', 'probability': 0.9980997810851449}]}]
233 [{'全部责任[正,负]': [{'text': '负', 'probability': 0.9526596587877592}]}]
234 [{'全部责任[正,负]': [{'text': '负', 'probability': 0.9992012760778124}]}]
235 [{'全部责任[正,负]': [{'text': '负', 'probability': 0.9992319018799378}]}]
236 [{'全部责任[正,负]': [{'text': '负', 'probability': 0.9993070334314886}]}]
task_name: 交通肇事后逃逸
[2022-10-20 22:57:50,317] [ INFO] - We are using <class 'paddlenlp.transformers.ernie.tokenizer.ErnieTokenizer'> to load 'checkpoint/model_best'.
187 [{'交通肇事后逃逸[正,负]': [{'text': '负', 'probability': 0.9989317097630845}]}]
188 [{'交通肇事后逃逸[正,负]': [{'text': '正', 'probability': 0.9993860236928356}]}]
189 [{'交通肇事后逃逸[正,负]': [{'text': '正', 'probability': 0.9996634405664793}]}]
190 [{'交通肇事后逃逸[正,负]': [{'text': '正', 'probability': 0.9992484074939654}]}]
191 [{'交通肇事后逃逸[正,负]': [{'text': '正', 'probability': 0.9953392118866233}]}]
192 [{'交通肇事后逃逸[正,负]': [{'text': '正', 'probability': 0.9995160553090159}]}]
193 [{'交通肇事后逃逸[正,负]': [{'text': '正', 'probability': 0.9995410959309758}]}]
194 [{'交通肇事后逃逸[正,负]': [{'text': '负', 'probability': 0.9903837969524432}]}]
195 [{'交通肇事后逃逸[正,负]': [{'text': '负', 'probability': 0.9609727115601245}]}]
196 [{'交通肇事后逃逸[正,负]': [{'text': '负', 'probability': 0.9968648619421749}]}]
197 [{'交通肇事后逃逸[正,负]': [{'text': '负', 'probability': 0.9991533523864504}]}]
198 [{'交通肇事后逃逸[正,负]': [{'text': '负', 'probability': 0.9985925638518012}]}]
199 [{'交通肇事后逃逸[正,负]': [{'text': '负', 'probability': 0.9596972444036425}]}]
200 [{'交通肇事后逃逸[正,负]': [{'text': '负', 'probability': 0.9988057392883611}]}]
201 [{'交通肇事后逃逸[正,负]': [{'text': '负', 'probability': 0.9992429723612162}]}]
202 [{'交通肇事后逃逸[正,负]': [{'text': '负', 'probability': 0.9991362034479678}]}]
203 [{'交通肇事后逃逸[正,负]': [{'text': '负', 'probability': 0.9992302176196475}]}]
204 [{'交通肇事后逃逸[正,负]': [{'text': '负', 'probability': 0.9992603728000304}]}]
205 [{'交通肇事后逃逸[正,负]': [{'text': '负', 'probability': 0.9992085325383044}]}]
206 [{'交通肇事后逃逸[正,负]': [{'text': '负', 'probability': 0.9990872881797515}]}]
207 [{'交通肇事后逃逸[正,负]': [{'text': '负', 'probability': 0.9989899936657665}]}]
208 [{'交通肇事后逃逸[正,负]': [{'text': '负', 'probability': 0.9991583000087978}]}]
209 [{'交通肇事后逃逸[正,负]': [{'text': '负', 'probability': 0.998197025698019}]}]
210 [{'交通肇事后逃逸[正,负]': [{'text': '负', 'probability': 0.9992022707045045}]}]
211 [{'交通肇事后逃逸[正,负]': [{'text': '负', 'probability': 0.9991474610293416}]}]
212 [{'交通肇事后逃逸[正,负]': [{'text': '负', 'probability': 0.9990975857623852}]}]
task_name: 肇事车辆超速行驶
[2022-10-20 22:57:52,751] [ INFO] - We are using <class 'paddlenlp.transformers.ernie.tokenizer.ErnieTokenizer'> to load 'checkpoint/model_best'.
237 [{'肇事车辆超速行驶[正,负]': [{'text': '负', 'probability': 0.9841847129404613}]}]
238 [{'肇事车辆超速行驶[正,负]': [{'text': '负', 'probability': 0.9856382489803082}]}]
239 [{'肇事车辆超速行驶[正,负]': [{'text': '正', 'probability': 0.9942912518391509}]}]
240 [{'肇事车辆超速行驶[正,负]': [{'text': '负', 'probability': 0.9990497924227171}]}]
241 [{'肇事车辆超速行驶[正,负]': [{'text': '负', 'probability': 0.9984236281120928}]}]
242 [{'肇事车辆超速行驶[正,负]': [{'text': '负', 'probability': 0.9797013417350584}]}]
243 [{'肇事车辆超速行驶[正,负]': [{'text': '负', 'probability': 0.9405926761127503}]}]
244 [{'肇事车辆超速行驶[正,负]': [{'text': '负', 'probability': 0.9973949246296403}]}]
245 [{'肇事车辆超速行驶[正,负]': [{'text': '负', 'probability': 0.9989976162783094}]}]
246 [{'肇事车辆超速行驶[正,负]': [{'text': '负', 'probability': 0.7694760238243088}]}]
247 [{'肇事车辆超速行驶[正,负]': [{'text': '负', 'probability': 0.9502718808871418}]}]
248 [{'肇事车辆超速行驶[正,负]': [{'text': '负', 'probability': 0.9957582481010405}]}]
249 [{'肇事车辆超速行驶[正,负]': [{'text': '正', 'probability': 0.9990645412108741}]}]
250 [{'肇事车辆超速行驶[正,负]': [{'text': '负', 'probability': 0.9983027472627981}]}]
251 [{'肇事车辆超速行驶[正,负]': [{'text': '负', 'probability': 0.9989368570279922}]}]
252 [{'肇事车辆超速行驶[正,负]': [{'text': '负', 'probability': 0.9989502518674271}]}]
253 [{'肇事车辆超速行驶[正,负]': [{'text': '负', 'probability': 0.9989551310134175}]}]
254 [{'肇事车辆超速行驶[正,负]': [{'text': '负', 'probability': 0.9990440336154052}]}]
255 [{'肇事车辆超速行驶[正,负]': [{'text': '负', 'probability': 0.9988904299118673}]}]
256 [{'肇事车辆超速行驶[正,负]': [{'text': '负', 'probability': 0.9989018226745472}]}]
257 [{'肇事车辆超速行驶[正,负]': [{'text': '负', 'probability': 0.9989507468908094}]}]
258 [{'肇事车辆超速行驶[正,负]': [{'text': '负', 'probability': 0.9988409186153646}]}]
259 [{'肇事车辆超速行驶[正,负]': [{'text': '负', 'probability': 0.9979641056034687}]}]
task_name: 被害人为本车人员
[2022-10-20 22:57:55,113] [ INFO] - We are using <class 'paddlenlp.transformers.ernie.tokenizer.ErnieTokenizer'> to load 'checkpoint/model_best'.
287 [{'被害人为本车人员[正,负]': [{'text': '正', 'probability': 0.9436778034352198}]}]
288 [{'被害人为本车人员[正,负]': [{'text': '正', 'probability': 0.9842731389150536}]}]
289 [{'被害人为本车人员[正,负]': [{'text': '正', 'probability': 0.999351125550529}]}]
290 [{'被害人为本车人员[正,负]': [{'text': '负', 'probability': 0.9970759981241173}]}]
291 [{'被害人为本车人员[正,负]': [{'text': '负', 'probability': 0.9990002419041701}]}]
292 [{'被害人为本车人员[正,负]': [{'text': '负', 'probability': 0.9931749358234896}]}]
293 [{'被害人为本车人员[正,负]': [{'text': '负', 'probability': 0.9935532557581261}]}]
294 [{'被害人为本车人员[正,负]': [{'text': '正', 'probability': 0.47645317553218547}]}]
295 [{'被害人为本车人员[正,负]': [{'text': '正', 'probability': 0.9940586172795491}]}]
296 [{'被害人为本车人员[正,负]': [{'text': '负', 'probability': 0.9991703323046188}]}]
297 [{'被害人为本车人员[正,负]': [{'text': '负', 'probability': 0.9744459753187567}]}]
298 [{'被害人为本车人员[正,负]': [{'text': '负', 'probability': 0.9994108236213073}]}]
299 [{'被害人为本车人员[正,负]': [{'text': '正', 'probability': 0.9257994065754147}]}]
300 [{'被害人为本车人员[正,负]': [{'text': '负', 'probability': 0.999024886453725}]}]
301 [{'被害人为本车人员[正,负]': [{'text': '负', 'probability': 0.9993531664113835}]}]
302 [{'被害人为本车人员[正,负]': [{'text': '负', 'probability': 0.8173563912408337}]}]
303 [{'被害人为本车人员[正,负]': [{'text': '负', 'probability': 0.7356733007322713}]}]
304 [{'被害人为本车人员[正,负]': [{'text': '正', 'probability': 0.9541713746476894}]}]
305 [{'被害人为本车人员[正,负]': [{'text': '负', 'probability': 0.9985592266450484}]}]
306 [{'被害人为本车人员[正,负]': [{'text': '负', 'probability': 0.999286466075155}]}]
307 [{'被害人为本车人员[正,负]': [{'text': '正', 'probability': 0.7782722899948915}]}]
308 [{'被害人为本车人员[正,负]': [{'text': '负', 'probability': 0.9992409504117319}]}]
309 [{'被害人为本车人员[正,负]': [{'text': '负', 'probability': 0.9992083718473008}]}]
310 [{'被害人为本车人员[正,负]': [{'text': '负', 'probability': 0.7672187500992086}]}]
311 [{'被害人为本车人员[正,负]': [{'text': '负', 'probability': 0.9992408637224237}]}]
312 [{'被害人为本车人员[正,负]': [{'text': '负', 'probability': 0.9985595806067149}]}]
313 [{'被害人为本车人员[正,负]': [{'text': '负', 'probability': 0.999301191831254}]}]
314 [{'被害人为本车人员[正,负]': [{'text': '负', 'probability': 0.9988617505389179}]}]
315 [{'被害人为本车人员[正,负]': [{'text': '负', 'probability': 0.9991491126623977}]}]
316 [{'被害人为本车人员[正,负]': [{'text': '负', 'probability': 0.9994118963302974}]}]
317 [{'被害人为本车人员[正,负]': [{'text': '正', 'probability': 0.9993650678817438}]}]
task_name: 中型客车交通肇事
[2022-10-20 22:57:58,566] [ INFO] - We are using <class 'paddlenlp.transformers.ernie.tokenizer.ErnieTokenizer'> to load 'checkpoint/model_best'.
160 [{'中型客车交通肇事[正,负]': [{'text': '正', 'probability': 0.9996376957524049}]}]
161 [{'中型客车交通肇事[正,负]': [{'text': '正', 'probability': 0.9994045572659545}]}]
162 [{'中型客车交通肇事[正,负]': [{'text': '正', 'probability': 0.999352190713541}]}]
163 [{'中型客车交通肇事[正,负]': [{'text': '正', 'probability': 0.9986186508948194}]}]
164 [{'中型客车交通肇事[正,负]': [{'text': '负', 'probability': 0.9985135229913098}]}]
165 [{'中型客车交通肇事[正,负]': [{'text': '正', 'probability': 0.9966814844411331}]}]
166 [{'中型客车交通肇事[正,负]': [{'text': '正', 'probability': 0.9995706527736381}]}]
167 [{'中型客车交通肇事[正,负]': [{'text': '负', 'probability': 0.9970932608194136}]}]
168 [{'中型客车交通肇事[正,负]': [{'text': '正', 'probability': 0.9993670818215428}]}]
169 [{'中型客车交通肇事[正,负]': [{'text': '负', 'probability': 0.9991417213458812}]}]
170 [{'中型客车交通肇事[正,负]': [{'text': '负', 'probability': 0.9983698894467725}]}]
171 [{'中型客车交通肇事[正,负]': [{'text': '负', 'probability': 0.9990088713607292}]}]
172 [{'中型客车交通肇事[正,负]': [{'text': '负', 'probability': 0.9993976620434921}]}]
173 [{'中型客车交通肇事[正,负]': [{'text': '负', 'probability': 0.9990538850322608}]}]
174 [{'中型客车交通肇事[正,负]': [{'text': '负', 'probability': 0.9994110103184184}]}]
175 [{'中型客车交通肇事[正,负]': [{'text': '负', 'probability': 0.9993443234750607}]}]
176 [{'中型客车交通肇事[正,负]': [{'text': '负', 'probability': 0.9988959478010528}]}]
177 [{'中型客车交通肇事[正,负]': [{'text': '负', 'probability': 0.9958639936573892}]}]
178 [{'中型客车交通肇事[正,负]': [{'text': '负', 'probability': 0.9995017034061888}]}]
179 [{'中型客车交通肇事[正,负]': [{'text': '负', 'probability': 0.9993751377443942}]}]
180 [{'中型客车交通肇事[正,负]': [{'text': '负', 'probability': 0.9994429458940246}]}]
181 [{'中型客车交通肇事[正,负]': [{'text': '负', 'probability': 0.9992461892294884}]}]
182 [{'中型客车交通肇事[正,负]': [{'text': '负', 'probability': 0.998882940724509}]}]
183 [{'中型客车交通肇事[正,负]': [{'text': '负', 'probability': 0.9993437258224667}]}]
184 [{'中型客车交通肇事[正,负]': [{'text': '负', 'probability': 0.9986758617482714}]}]
185 [{'中型客车交通肇事[正,负]': [{'text': '负', 'probability': 0.9990830735000102}]}]
186 [{'中型客车交通肇事[正,负]': [{'text': '负', 'probability': 0.999167480901864}]}]
task_name: ner
[2022-10-20 22:58:01,028] [ INFO] - We are using <class 'paddlenlp.transformers.ernie.tokenizer.ErnieTokenizer'> to load 'checkpoint/model_best'.
[2022-10-20 22:58:02,577] [ INFO] - We are using <class 'paddlenlp.transformers.ernie.tokenizer.ErnieTokenizer'> to load 'checkpoint/model_best'.
0 [{'被害人': [{'text': '谌某某', 'start': 160, 'end': 163, 'probability': 0.9999530319317529}, {'text': '韩某某', 'start': 164, 'end': 167, 'probability': 0.9999778271926516}, {'text': '王某某', 'start': 156, 'end': 159, 'probability': 0.9999740125391128}, {'text': '韩某某', 'start': 42, 'end': 45, 'probability': 0.9999816418536511}, {'text': '王某某', 'start': 38, 'end': 41, 'probability': 0.9999727012430526}, {'text': '黄某', 'start': 46, 'end': 48, 'probability': 0.9999668600620026}, {'text': '朱某某', 'start': 172, 'end': 175, 'probability': 0.9998762645852395}, {'text': '黄某', 'start': 176, 'end': 178, 'probability': 0.9999363432330028}, {'text': '朱某某', 'start': 92, 'end': 95, 'probability': 0.9997549207065362}], '犯罪嫌疑人交通工具': [{'text': '大型普通客车', 'start': 31, 'end': 37, 'probability': 0.999982476305803}], '被害人交通工具情况': [{'text': '无牌', 'start': 98, 'end': 100, 'probability': 0.99988401272131}], '事故发生地': [{'text': '道真自治县上坝乡双河村谢村沟路段', 'start': 71, 'end': 87, 'probability': 0.9999645951374703}], '被害人交通工具': [{'text': '二轮摩托车', 'start': 100, 'end': 105, 'probability': 0.999961376403462}]}]
1 [{'犯罪嫌疑人情况': [{'text': '酒后', 'start': 32, 'end': 34, 'probability': 0.9999810458068055}], '被害人类型': [{'text': '行人', 'start': 96, 'end': 98, 'probability': 0.999736611734761}], '犯罪嫌疑人交通工具': [{'text': '电动自行车', 'start': 45, 'end': 50, 'probability': 0.9999822378736667}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 175, 'end': 179, 'probability': 0.9999787808061456}], '事故发生地': [{'text': '雁江区雁江镇王雁路1KM+200M处', 'start': 72, 'end': 90, 'probability': 0.9999672176895729}]}]
2 [{'被害人': [{'text': '曾某', 'start': 76, 'end': 78, 'probability': 0.9999558930403509}, {'text': '曾某', 'start': 69, 'end': 71, 'probability': 0.9999827147271105}], '犯罪嫌疑人交通工具': [{'text': '小型轿车', 'start': 31, 'end': 35, 'probability': 0.9999620918031411}], '犯罪嫌疑人责任认定': [{'text': '主要责任', 'start': 121, 'end': 125, 'probability': 0.9999722243963447}], '事故发生地': [{'text': '高速交警大队路段', 'start': 46, 'end': 54, 'probability': 0.9999468333546702}]}]
3 [{'被害人': [{'text': '王某', 'start': 85, 'end': 87, 'probability': 0.9999789000668784}, {'text': '王某', 'start': 99, 'end': 101, 'probability': 0.9999748469979011}, {'text': '陶某', 'start': 104, 'end': 106, 'probability': 0.9999516015253676}, {'text': '陶某', 'start': 88, 'end': 90, 'probability': 0.9999828339323358}], '被害人类型': [{'text': '行人', 'start': 83, 'end': 85, 'probability': 0.9999623302181107}, {'text': '行人', 'start': 97, 'end': 99, 'probability': 0.9998918799831813}], '犯罪嫌疑人交通工具': [{'text': '小型轿车', 'start': 29, 'end': 33, 'probability': 0.9999738933065601}], '事故发生地': [{'text': '上松线48KM+300M地段', 'start': 51, 'end': 65, 'probability': 0.9999710323242823}]}]
4 [{'犯罪嫌疑人情况': [{'text': '酒后', 'start': 30, 'end': 32, 'probability': 0.9996697928040703}, {'text': '无证', 'start': 32, 'end': 34, 'probability': 0.9999676945284364}], '被害人': [{'text': '刘某某', 'start': 97, 'end': 100, 'probability': 0.9999870062277694}, {'text': '刘某某', 'start': 89, 'end': 92, 'probability': 0.999990940112113}], '被害人类型': [{'text': '行人', 'start': 87, 'end': 89, 'probability': 0.9999706747208847}], '犯罪嫌疑人交通工具': [{'text': '小型轿车', 'start': 44, 'end': 48, 'probability': 0.9999856949360577}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 120, 'end': 124, 'probability': 0.9999752045595756}], '事故发生地': [{'text': '兴隆县大苇塘路段', 'start': 60, 'end': 68, 'probability': 0.7649837672858268}]}]
5 [{'被害人': [{'text': '陈某', 'start': 16, 'end': 18, 'probability': 0.9999663832627448}, {'text': '李某', 'start': 19, 'end': 21, 'probability': 0.9999635222899315}, {'text': '陈某', 'start': 237, 'end': 239, 'probability': 0.9999459989149102}, {'text': '陈某', 'start': 216, 'end': 218, 'probability': 0.9999189392582224}, {'text': '李某', 'start': 312, 'end': 314, 'probability': 0.9997648132415122}, {'text': '李某', 'start': 249, 'end': 251, 'probability': 0.9992097731546714}], '被害人类型': [{'text': '乘车人', 'start': 234, 'end': 237, 'probability': 0.9876105510415663}, {'text': '乘车人', 'start': 246, 'end': 249, 'probability': 0.8049189055257244}], '犯罪嫌疑人交通工具': [{'text': '小型普通客车', 'start': 129, 'end': 135, 'probability': 0.9999837875977278}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 300, 'end': 304, 'probability': 0.9999798537174343}], '事故发生地': [{'text': '中牟县土寨道北031号线杆路口处', 'start': 154, 'end': 170, 'probability': 0.9999172702696342}]}]
6 [{'被害人': [{'text': '于某2', 'start': 219, 'end': 222, 'probability': 0.9999581579284325}, {'text': '于某2', 'start': 163, 'end': 166, 'probability': 0.9999870062262346}, {'text': '于某2', 'start': 135, 'end': 138, 'probability': 0.9999881983151226}], '被害人类型': [{'text': '乘坐人', 'start': 132, 'end': 135, 'probability': 0.9999822378939314}], '犯罪嫌疑人交通工具': [{'text': '重型自卸货车', 'start': 34, 'end': 40, 'probability': 0.9999786616494362}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 211, 'end': 215, 'probability': 0.9999588731859035}], '被害人责任认定': [{'text': '无事故责任', 'start': 222, 'end': 227, 'probability': 0.9999827147066469}], '事故发生地': [{'text': '万官大街傲徕峰路口', 'start': 55, 'end': 64, 'probability': 0.9978358699724552}], '被害人交通工具': [{'text': '电动二轮车', 'start': 107, 'end': 112, 'probability': 0.9999835491672826}]}]
7 [{'被害人': [{'text': '高某某', 'start': 110, 'end': 113, 'probability': 0.999960303673646}, {'text': '陈某某', 'start': 140, 'end': 143, 'probability': 0.9999914169478501}, {'text': '高某某', 'start': 128, 'end': 131, 'probability': 0.9999864101851159}, {'text': '高某某', 'start': 167, 'end': 170, 'probability': 0.9999825955039796}], '被害人类型': [{'text': '乘车人', 'start': 137, 'end': 140, 'probability': 0.999958396286786}], '犯罪嫌疑人交通工具': [{'text': '轻型自卸车', 'start': 36, 'end': 41, 'probability': 0.9999848604741146}], '犯罪嫌疑人责任认定': [{'text': '主要责任', 'start': 266, 'end': 270, 'probability': 0.9999741317257929}], '事故发生地': [{'text': '涡阳县涡南镇王塘村路段', 'start': 71, 'end': 82, 'probability': 0.9999691250041849}], '被害人交通工具': [{'text': '电动轮车', 'start': 119, 'end': 123, 'probability': 0.9999834299505039}]}]
8 [{'犯罪嫌疑人情况': [{'text': '无证', 'start': 29, 'end': 31, 'probability': 0.9999524358492806}], '被害人': [{'text': '曹某', 'start': 182, 'end': 184, 'probability': 0.999972701257434}, {'text': '曹某', 'start': 88, 'end': 90, 'probability': 0.9999749662026716}, {'text': '曹某', 'start': 209, 'end': 211, 'probability': 0.9978399785042456}, {'text': '曹某', 'start': 106, 'end': 108, 'probability': 0.9999742509523344}], '犯罪嫌疑人交通工具': [{'text': '自卸三轮汽车', 'start': 38, 'end': 44, 'probability': 0.9884338701782838}], '犯罪嫌疑人交通工具情况': [{'text': '无号牌', 'start': 33, 'end': 36, 'probability': 0.9999712707155481}], '被害人交通工具情况': [{'text': '无号牌', 'start': 93, 'end': 96, 'probability': 0.9999434955565363}], '犯罪嫌疑人责任认定': [{'text': '主要责任', 'start': 151, 'end': 155, 'probability': 0.9999721052013228}], '事故发生地': [{'text': '翟洼村松楼组处路段', 'start': 71, 'end': 80, 'probability': 0.9999734164794347}], '被害人交通工具': [{'text': '二轮摩托车', 'start': 96, 'end': 101, 'probability': 0.9999861717528091}]}]
9 [{'被害人': [{'text': '秦某某', 'start': 424, 'end': 427, 'probability': 0.9999802113342184}, {'text': '李某丙', 'start': 170, 'end': 173, 'probability': 0.9999876022691154}, {'text': '秦某某', 'start': 166, 'end': 169, 'probability': 0.999989867234774}, {'text': '李某甲', 'start': 156, 'end': 159, 'probability': 0.9999892711912963}, {'text': '李某甲', 'start': 255, 'end': 258, 'probability': 0.9999850988920969}, {'text': '李某丙', 'start': 277, 'end': 280, 'probability': 0.9999800921398787}, {'text': '张某某', 'start': 577, 'end': 580, 'probability': 0.9912750718748953}, {'text': '李某丙', 'start': 623, 'end': 626, 'probability': 0.99997222442056}, {'text': '李某甲', 'start': 615, 'end': 618, 'probability': 0.9999793768992049}, {'text': '刘某乙', 'start': 573, 'end': 576, 'probability': 0.9634757078321243}, {'text': '陈某某', 'start': 627, 'end': 630, 'probability': 0.9999732972937636}, {'text': '秦某某', 'start': 619, 'end': 622, 'probability': 0.9999800921438577}], '被害人类型': [{'text': '乘员', 'start': 154, 'end': 156, 'probability': 0.9999672177110597}, {'text': '乘员', 'start': 164, 'end': 166, 'probability': 0.999949694238822}], '犯罪嫌疑人交通工具': [{'text': '出租车', 'start': 39, 'end': 42, 'probability': 0.9999630454615271}], '犯罪嫌疑人责任认定': [{'text': '主要责任', 'start': 568, 'end': 572, 'probability': 0.9999787808362868}], '被害人责任认定': [{'text': '无责任', 'start': 630, 'end': 633, 'probability': 0.9999798537266713}], '事故发生地': [{'text': '明沈公路123KM+800M处', 'start': 50, 'end': 65, 'probability': 0.9998475338356627}], '被害人交通工具': [{'text': '小型轿车', 'start': 91, 'end': 95, 'probability': 0.995012144919798}, {'text': '重型半挂牵引车', 'start': 130, 'end': 137, 'probability': 0.9998565960485735}]}]
10 [{'犯罪嫌疑人情况': [{'text': '无驾驶证', 'start': 30, 'end': 34, 'probability': 0.9999084480130023}], '被害人': [{'text': '卜某某', 'start': 74, 'end': 77, 'probability': 0.9999884367310585}, {'text': '卜某某', 'start': 95, 'end': 98, 'probability': 0.9999859333521499}], '犯罪嫌疑人交通工具': [{'text': '三轮汽车', 'start': 38, 'end': 42, 'probability': 0.9998739986949658}], '犯罪嫌疑人交通工具情况': [{'text': '无牌', 'start': 36, 'end': 38, 'probability': 0.9999177469223923}], '犯罪嫌疑人责任认定': [{'text': '主要责任', 'start': 120, 'end': 124, 'probability': 0.9999721051893857}], '事故发生地': [{'text': '东平县东平某政府东豆山村南侧', 'start': 50, 'end': 64, 'probability': 0.9999692442392671}], '被害人交通工具': [{'text': '电动三轮车', 'start': 80, 'end': 85, 'probability': 0.999983668377098}]}]
11 [{'被害人': [{'text': '王某', 'start': 169, 'end': 171, 'probability': 0.9999818802631921}, {'text': '王某', 'start': 119, 'end': 121, 'probability': 0.9999548201489006}, {'text': '王某', 'start': 103, 'end': 105, 'probability': 0.999979853731503}, {'text': '王某', 'start': 340, 'end': 342, 'probability': 0.8919530114217977}, {'text': '王某', 'start': 145, 'end': 147, 'probability': 0.9999611381491604}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 325, 'end': 329, 'probability': 0.999984622051528}], '事故发生地': [{'text': '与巴彦路交叉路口', 'start': 70, 'end': 78, 'probability': 0.999039427308368}]}]
12 [{'被害人': [{'text': '王某某', 'start': 356, 'end': 359, 'probability': 0.9999679329575883}, {'text': '王某某', 'start': 222, 'end': 225, 'probability': 0.9999722244283191}, {'text': '王某某', 'start': 301, 'end': 304, 'probability': 0.9998513507919427}, {'text': '王某某', 'start': 126, 'end': 129, 'probability': 0.9999711515154246}, {'text': '王某某', 'start': 74, 'end': 77, 'probability': 0.9999877214786466}, {'text': '王某某', 'start': 191, 'end': 194, 'probability': 0.999981045802798}], '犯罪嫌疑人交通工具': [{'text': '小型轿车', 'start': 36, 'end': 40, 'probability': 0.9999792576879827}], '犯罪嫌疑人责任认定': [{'text': '主要责任', 'start': 182, 'end': 186, 'probability': 0.9999692442261221}], '事故发生地': [{'text': '李村加油站路段', 'start': 54, 'end': 61, 'probability': 0.9779995199453069}], '被害人交通工具': [{'text': '轻型普通货车', 'start': 108, 'end': 114, 'probability': 0.9998893768018746}]}]
13 [{'被害人': [{'text': '何某', 'start': 169, 'end': 171, 'probability': 0.9999789000644626}, {'text': '何某', 'start': 84, 'end': 86, 'probability': 0.9999791384752825}, {'text': '何某', 'start': 195, 'end': 197, 'probability': 0.9999704363146975}, {'text': '何某', 'start': 148, 'end': 150, 'probability': 0.9999783040233439}], '犯罪嫌疑人交通工具': [{'text': '小型轿车', 'start': 38, 'end': 42, 'probability': 0.9999738933329922}], '事故发生地': [{'text': '七雄路某公司路段', 'start': 70, 'end': 78, 'probability': 0.9998281038519963}, {'text': '本市东西湖区七雄路', 'start': 44, 'end': 53, 'probability': 0.6619407851411836}], '被害人交通工具': [{'text': '两轮电动车', 'start': 97, 'end': 102, 'probability': 0.9999792576383015}]}]
14 [{'犯罪嫌疑人情况': [{'text': '无有效驾驶证', 'start': 39, 'end': 45, 'probability': 0.9998499210569207}], '被害人': [{'text': '蒋某某', 'start': 114, 'end': 117, 'probability': 0.9999901056515625}, {'text': '蒋某某', 'start': 106, 'end': 109, 'probability': 0.9999901056531542}], '被害人类型': [{'text': '行人', 'start': 104, 'end': 106, 'probability': 0.999983310767675}], '犯罪嫌疑人交通工具': [{'text': '正三轮摩托车', 'start': 32, 'end': 38, 'probability': 0.999976158282152}], '犯罪嫌疑人交通工具情况': [{'text': '无号牌', 'start': 29, 'end': 32, 'probability': 0.9999184616758612}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 258, 'end': 262, 'probability': 0.9999530319435763}], '事故发生地': [{'text': '国道212线1150KM+500M处', 'start': 74, 'end': 92, 'probability': 0.9999655487129644}]}]
15 [{'犯罪嫌疑人情况': [{'text': '超速', 'start': 118, 'end': 120, 'probability': 0.9991690925561869}], '被害人': [{'text': '王某3', 'start': 134, 'end': 137, 'probability': 0.99998486047744}, {'text': '王某3', 'start': 246, 'end': 249, 'probability': 0.9999750854134959}, {'text': '王某3', 'start': 202, 'end': 205, 'probability': 0.9999797345171828}, {'text': '王某3', 'start': 76, 'end': 79, 'probability': 0.9999892711918079}], '犯罪嫌疑人交通工具': [{'text': '吉利轿车', 'start': 36, 'end': 40, 'probability': 0.9997704704186248}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 238, 'end': 242, 'probability': 0.9999102372489972}], '被害人责任认定': [{'text': '不承担事故责任', 'start': 249, 'end': 256, 'probability': 0.9999734164283325}], '事故发生地': [{'text': '天津市西青区杨柳青镇莱茵小镇小区门口', 'start': 52, 'end': 70, 'probability': 0.9999715091533119}]}]
16 [{'被害人': [{'text': '赵某某', 'start': 163, 'end': 166, 'probability': 0.9999864101796447}, {'text': '赵某某', 'start': 74, 'end': 77, 'probability': 0.9999899864384787}, {'text': '赵某某', 'start': 91, 'end': 94, 'probability': 0.9999910593226673}, {'text': '赵某某', 'start': 114, 'end': 117, 'probability': 0.9999895096031537}], '被害人类型': [{'text': '行人', 'start': 72, 'end': 74, 'probability': 0.9999779464017422}], '犯罪嫌疑人交通工具': [{'text': '轻型厢式货车', 'start': 38, 'end': 44, 'probability': 0.9999837875999305}], '犯罪嫌疑人责任认定': [{'text': '主要责任', 'start': 158, 'end': 162, 'probability': 0.9999837876022042}], '被害人责任认定': [{'text': '次要责任', 'start': 171, 'end': 175, 'probability': 0.999964714263669}], '事故发生地': [{'text': '沈阳市于洪区阳光路吉力湖一街路口西500米处', 'start': 47, 'end': 69, 'probability': 0.9999741317514577}]}]
17 [{'被害人': [{'text': '姬某2', 'start': 87, 'end': 90, 'probability': 0.9999904632784222}, {'text': '姬某2', 'start': 102, 'end': 105, 'probability': 0.9999817610596438}, {'text': '姬某2', 'start': 117, 'end': 120, 'probability': 0.9999563698102207}], '犯罪嫌疑人交通工具': [{'text': '轻型普通货车', 'start': 34, 'end': 40, 'probability': 0.999985456517507}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 193, 'end': 197, 'probability': 0.9999781847634779}], '事故发生地': [{'text': '梁某某—滩郭011号线杆处', 'start': 63, 'end': 76, 'probability': 0.9999364620496749}]}]
18 [{'被害人': [{'text': '刘某甲', 'start': 116, 'end': 119, 'probability': 0.9999871254340604}, {'text': '刘某甲', 'start': 134, 'end': 137, 'probability': 0.9999868870148703}, {'text': '刘某甲', 'start': 143, 'end': 146, 'probability': 0.9999870062230798}], '被害人类型': [{'text': '行人', 'start': 111, 'end': 113, 'probability': 0.9999490981935537}], '犯罪嫌疑人交通工具': [{'text': '低速自卸货车', 'start': 49, 'end': 55, 'probability': 0.999757828679499}], '事故发生地': [{'text': '龙门县永汉镇永强学校路口路段', 'start': 83, 'end': 97, 'probability': 0.9999791384699108}]}]
19 [{'犯罪嫌疑人情况': [{'text': '超速', 'start': 87, 'end': 89, 'probability': 0.8012725884402983}, {'text': '超载', 'start': 84, 'end': 86, 'probability': 0.5433102473995746}], '被害人': [{'text': '孟某2', 'start': 110, 'end': 113, 'probability': 0.9999777079785588}, {'text': '孟某2', 'start': 140, 'end': 143, 'probability': 0.9991321746318675}, {'text': '孟某2', 'start': 123, 'end': 126, 'probability': 0.9951032946509031}, {'text': '孟某2', 'start': 177, 'end': 180, 'probability': 0.9987934537009799}], '被害人类型': [{'text': '行人', 'start': 108, 'end': 110, 'probability': 0.9999767543192775}], '犯罪嫌疑人交通工具': [{'text': '重型半挂牵引车', 'start': 40, 'end': 47, 'probability': 0.9999845028461323}], '犯罪嫌疑人责任认定': [{'text': '主要责任', 'start': 172, 'end': 176, 'probability': 0.9999746085387358}], '被害人责任认定': [{'text': '次要责任', 'start': 182, 'end': 186, 'probability': 0.9999668600078166}], '事故发生地': [{'text': '大连市普兰店区S314线', 'start': 48, 'end': 60, 'probability': 0.9999308592387592}, {'text': '241km+138m处', 'start': 69, 'end': 80, 'probability': 0.9006947566980728}]}]
20 [{'犯罪嫌疑人情况': [{'text': '未年检', 'start': 32, 'end': 35, 'probability': 0.9994537845190337}], '被害人': [{'text': '刘某1', 'start': 98, 'end': 101, 'probability': 0.9999881983117405}, {'text': '刘某1', 'start': 108, 'end': 111, 'probability': 0.9999852180923625}], '被害人类型': [{'text': '行人', 'start': 96, 'end': 98, 'probability': 0.9999778271899231}], '犯罪嫌疑人交通工具': [{'text': '中型厢式货车', 'start': 42, 'end': 48, 'probability': 0.9999513629266943}], '犯罪嫌疑人交通工具情况': [{'text': '未年检', 'start': 32, 'end': 35, 'probability': 0.9999520782889704}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 178, 'end': 182, 'probability': 0.9999744893090252}], '事故发生地': [{'text': '105国道赣州市南康区龙回镇半岭村路段', 'start': 67, 'end': 86, 'probability': 0.9999669792945269}]}]
21 [{'犯罪嫌疑人情况': [{'text': '无证', 'start': 27, 'end': 29, 'probability': 0.9999644758216419}], '被害人': [{'text': '周某1', 'start': 115, 'end': 118, 'probability': 0.999958634803022}, {'text': '周某1', 'start': 99, 'end': 102, 'probability': 0.999989628817417}, {'text': '周某1', 'start': 129, 'end': 132, 'probability': 0.9999833107663108}], '犯罪嫌疑人交通工具': [{'text': '正三轮摩托车', 'start': 43, 'end': 49, 'probability': 0.9999841452157909}], '犯罪嫌疑人交通工具情况': [{'text': '无号牌', 'start': 31, 'end': 34, 'probability': 0.9999780655847275}], '事故发生地': [{'text': '襄阳市樊城区“前贾某某菜市场”门前路段', 'start': 65, 'end': 84, 'probability': 0.9999754430206877}]}]
22 [{'被害人': [{'text': '唐某乙', 'start': 80, 'end': 83, 'probability': 0.9999912977401948}, {'text': '唐某乙', 'start': 92, 'end': 95, 'probability': 0.9999881983147247}, {'text': '唐某乙', 'start': 142, 'end': 145, 'probability': 0.9999673369167823}], '犯罪嫌疑人交通工具': [{'text': '小型轿车', 'start': 38, 'end': 42, 'probability': 0.9999809265994912}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 137, 'end': 141, 'probability': 0.9999789000459884}], '被害人责任认定': [{'text': '不承担责任', 'start': 145, 'end': 150, 'probability': 0.9999849796704439}], '事故发生地': [{'text': '邹城市营西路圣源面粉厂路口处', 'start': 50, 'end': 64, 'probability': 0.9991749003918713}], '被害人交通工具': [{'text': '自行车', 'start': 67, 'end': 70, 'probability': 0.9999716283441558}]}]
23 [{'被害人': [{'text': '王某某', 'start': 201, 'end': 204, 'probability': 0.999982118683306}, {'text': '王某某', 'start': 118, 'end': 121, 'probability': 0.9999687673605422}, {'text': '王某某', 'start': 281, 'end': 284, 'probability': 0.9999698402758668}, {'text': '王某某某', 'start': 342, 'end': 346, 'probability': 0.77009954565969}, {'text': '王某某', 'start': 143, 'end': 146, 'probability': 0.999988079105421}], '犯罪嫌疑人交通工具': [{'text': '重型仓栅式货车', 'start': 38, 'end': 45, 'probability': 0.9999804497720532}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 254, 'end': 258, 'probability': 0.9999694826435643}], '事故发生地': [{'text': '104KM处弯道', 'start': 58, 'end': 66, 'probability': 0.9977446185845658}, {'text': '306省道', 'start': 46, 'end': 51, 'probability': 0.9698441986410558}], '被害人交通工具': [{'text': '小型普通客车', 'start': 132, 'end': 138, 'probability': 0.9999849796835178}]}]
24 [{'被害人': [{'text': '马某3', 'start': 85, 'end': 88, 'probability': 0.9999873638528243}, {'text': '马某3', 'start': 77, 'end': 80, 'probability': 0.9999889135657725}, {'text': '马某3', 'start': 121, 'end': 124, 'probability': 0.9999145286519706}], '犯罪嫌疑人交通工具': [{'text': '小轿车', 'start': 41, 'end': 44, 'probability': 0.9999785424284369}], '犯罪嫌疑人责任认定': [{'text': '主要责任', 'start': 116, 'end': 120, 'probability': 0.9999285948987335}], '被害人责任认定': [{'text': '次要责任', 'start': 127, 'end': 131, 'probability': 0.999971985906825}], '事故发生地': [{'text': '560公里加500米南半幅', 'start': 58, 'end': 71, 'probability': 0.9996088486466377}, {'text': '连霍高速公路', 'start': 45, 'end': 51, 'probability': 0.999562566288219}]}]
25 [{'被害人': [{'text': '俞某', 'start': 170, 'end': 172, 'probability': 0.9999638798692274}, {'text': '俞某', 'start': 221, 'end': 223, 'probability': 0.9999673369145086}, {'text': '安某', 'start': 95, 'end': 97, 'probability': 0.9999458797127687}, {'text': '俞某', 'start': 67, 'end': 69, 'probability': 0.999954939383386}, {'text': '俞某', 'start': 124, 'end': 126, 'probability': 0.9999798537313609}, {'text': '安某', 'start': 129, 'end': 131, 'probability': 0.9999771119437355}, {'text': '安某', 'start': 224, 'end': 226, 'probability': 0.9997911560932948}], '犯罪嫌疑人交通工具': [{'text': '小型轿车', 'start': 32, 'end': 36, 'probability': 0.9999452834000522}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 199, 'end': 203, 'probability': 0.9999568466445368}], '事故发生地': [{'text': '漯河市郾城区某路', 'start': 37, 'end': 45, 'probability': 0.9961210885892058}, {'text': '某公司南门处', 'start': 52, 'end': 58, 'probability': 0.999632093531801}], '被害人交通工具': [{'text': '二轮电动车', 'start': 78, 'end': 83, 'probability': 0.9999837875941466}, {'text': '二轮电动车', 'start': 105, 'end': 110, 'probability': 0.9969565905519744}]}]
26 [{'被害人': [{'text': '苏某某', 'start': 102, 'end': 105, 'probability': 0.9999847412486247}, {'text': '苏某某', 'start': 116, 'end': 119, 'probability': 0.9999716283408588}, {'text': '苏某某', 'start': 244, 'end': 247, 'probability': 0.9999674560433789}, {'text': '苏某某', 'start': 93, 'end': 96, 'probability': 0.9999841452062697}], '被害人类型': [{'text': '行人', 'start': 91, 'end': 93, 'probability': 0.9999759198679641}], '犯罪嫌疑人交通工具': [{'text': '大型普通客车', 'start': 46, 'end': 52, 'probability': 0.9999873638524406}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 236, 'end': 240, 'probability': 0.9992915578651207}], '被害人责任认定': [{'text': '无责任', 'start': 247, 'end': 250, 'probability': 0.9999836683868182}], '事故发生地': [{'text': '稻香宾馆路段', 'start': 72, 'end': 78, 'probability': 0.9999495749419793}]}]
27 [{'被害人': [{'text': '吴某某', 'start': 73, 'end': 76, 'probability': 0.9999902248600989}], '被害人类型': [{'text': '行人', 'start': 71, 'end': 73, 'probability': 0.9999794961072439}], '犯罪嫌疑人交通工具': [{'text': '小型轿车', 'start': 32, 'end': 36, 'probability': 0.9999843836436071}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 228, 'end': 232, 'probability': 0.9999712706928676}], '事故发生地': [{'text': '红安县城关镇东上店路段', 'start': 53, 'end': 64, 'probability': 0.9999655488090582}]}]
28 [{'被害人': [{'text': '刘某某', 'start': 159, 'end': 162, 'probability': 0.9999791384631465}, {'text': '宋某', 'start': 110, 'end': 112, 'probability': 0.999978304026925}, {'text': '宋某', 'start': 200, 'end': 202, 'probability': 0.9999465949504156}, {'text': '宋某', 'start': 156, 'end': 158, 'probability': 0.9999766351148196}, {'text': '刘某某', 'start': 117, 'end': 120, 'probability': 0.9872131404817992}, {'text': '刘某某', 'start': 213, 'end': 216, 'probability': 0.999709685933766}], '犯罪嫌疑人交通工具': [{'text': '重型自卸货车', 'start': 40, 'end': 46, 'probability': 0.9999802113553216}], '犯罪嫌疑人责任认定': [{'text': '主要责任', 'start': 195, 'end': 199, 'probability': 0.9999790192534732}], '被害人责任认定': [{'text': '不承担责任', 'start': 216, 'end': 221, 'probability': 0.9999786616133122}], '事故发生地': [{'text': '广汉市和兴镇安平村旌江干道和兴路口', 'start': 68, 'end': 85, 'probability': 0.999957442733006}], '被害人交通工具': [{'text': '小型普通客车', 'start': 128, 'end': 134, 'probability': 0.9999845028359715}]}]
29 [{'被害人': [{'text': '冯某平', 'start': 200, 'end': 203, 'probability': 0.9999343167540928}, {'text': '冯某平', 'start': 140, 'end': 143, 'probability': 0.9999849796858626}, {'text': '冯某平', 'start': 105, 'end': 108, 'probability': 0.9999848604773547}], '犯罪嫌疑人交通工具': [{'text': '小型普通客车', 'start': 43, 'end': 49, 'probability': 0.9999837875999305}], '被害人交通工具情况': [{'text': '无号牌', 'start': 126, 'end': 129, 'probability': 0.9998049818827894}], '犯罪嫌疑人责任认定': [{'text': '主要责任', 'start': 195, 'end': 199, 'probability': 0.9999413496674805}], '被害人责任认定': [{'text': '次要责任', 'start': 210, 'end': 214, 'probability': 0.9999433761507248}], '事故发生地': [{'text': '省道205线362km+200m(三台县中新镇场镇路段)', 'start': 75, 'end': 103, 'probability': 0.9991797670526523}], '被害人交通工具': [{'text': '正三轮摩托车', 'start': 129, 'end': 135, 'probability': 0.9997184893840156}]}]
30 [{'被害人': [{'text': '董某', 'start': 94, 'end': 96, 'probability': 0.9992750290441847}, {'text': '谢某', 'start': 89, 'end': 91, 'probability': 0.9999679329503408}, {'text': '谢某', 'start': 108, 'end': 110, 'probability': 0.9999661448455299}, {'text': '董某', 'start': 83, 'end': 85, 'probability': 0.9999741317325288}, {'text': '谢某', 'start': 80, 'end': 82, 'probability': 0.9999772311493302}, {'text': '董某', 'start': 149, 'end': 151, 'probability': 0.9999638799054651}], '被害人类型': [{'text': '行人', 'start': 75, 'end': 77, 'probability': 0.999976158261461}], '犯罪嫌疑人交通工具': [{'text': '轻型厢式货车', 'start': 34, 'end': 40, 'probability': 0.9999840260188648}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 181, 'end': 185, 'probability': 0.9999825954978405}], '事故发生地': [{'text': '海拉尔区景观大道西侧道路', 'start': 41, 'end': 53, 'probability': 0.9999520783597973}, {'text': '17号灯杆西侧400米处', 'start': 60, 'end': 72, 'probability': 0.9999721051893857}]}]
31 [{'被害人': [{'text': '徐某某', 'start': 85, 'end': 88, 'probability': 0.999977350351287}, {'text': '徐某某', 'start': 106, 'end': 109, 'probability': 0.9999521975693995}, {'text': '徐某某', 'start': 169, 'end': 172, 'probability': 0.9999517206437929}], '犯罪嫌疑人交通工具': [{'text': '重型普通半挂车', 'start': 41, 'end': 48, 'probability': 0.9998931895978416}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 164, 'end': 168, 'probability': 0.9999406344952178}], '被害人责任认定': [{'text': '无责任', 'start': 172, 'end': 175, 'probability': 0.9999804497536502}], '事故发生地': [{'text': '唐通线丰润区口(12公里200米)', 'start': 50, 'end': 67, 'probability': 0.9999032033402244}], '被害人交通工具': [{'text': '电动两轮车', 'start': 91, 'end': 96, 'probability': 0.9999803305382784}]}]
32 [{'被害人': [{'text': '周XX', 'start': 137, 'end': 140, 'probability': 0.9999618533911274}, {'text': '周XX', 'start': 175, 'end': 178, 'probability': 0.9989143895513877}, {'text': '周XX', 'start': 101, 'end': 104, 'probability': 0.99990248893819}, {'text': '周XX', 'start': 108, 'end': 111, 'probability': 0.999672737369039}], '被害人类型': [{'text': '驾驶员', 'start': 134, 'end': 137, 'probability': 0.9999628070147537}], '犯罪嫌疑人交通工具': [{'text': '轻型普通货车', 'start': 39, 'end': 45, 'probability': 0.9999806881869091}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 235, 'end': 239, 'probability': 0.9999704362260076}], '事故发生地': [{'text': '双鸭山市尖山区隆安小区门前路口', 'start': 48, 'end': 63, 'probability': 0.9999643567400369}], '被害人交通工具': [{'text': '二轮摩托车', 'start': 117, 'end': 122, 'probability': 0.9999866485937758}]}]
33 [{'犯罪嫌疑人情况': [{'text': '超速', 'start': 138, 'end': 140, 'probability': 0.8601877157147086}], '被害人': [{'text': '黄某1', 'start': 61, 'end': 64, 'probability': 0.9999007004731482}, {'text': '田某', 'start': 58, 'end': 60, 'probability': 0.9996195757360411}, {'text': '黄某2', 'start': 271, 'end': 274, 'probability': 0.9999059460174351}, {'text': '王某', 'start': 193, 'end': 195, 'probability': 0.9999195347065495}, {'text': '田某', 'start': 264, 'end': 266, 'probability': 0.9997514628847171}, {'text': '黄某2', 'start': 203, 'end': 206, 'probability': 0.9999339590575147}, {'text': '杨某某', 'start': 207, 'end': 210, 'probability': 0.9998617214475303}, {'text': '黄某2', 'start': 65, 'end': 68, 'probability': 0.9999480254012241}, {'text': '王某', 'start': 261, 'end': 263, 'probability': 0.9988912644232073}, {'text': '黄某1', 'start': 267, 'end': 270, 'probability': 0.9997607599487424}, {'text': '田某', 'start': 196, 'end': 198, 'probability': 0.999937773645243}, {'text': '郑某某', 'start': 185, 'end': 188, 'probability': 0.9999626878350227}, {'text': '王某', 'start': 55, 'end': 57, 'probability': 0.9996375019307493}, {'text': '黄某1', 'start': 199, 'end': 202, 'probability': 0.9999089261537506}], '犯罪嫌疑人交通工具': [{'text': '小型普通客车', 'start': 40, 'end': 46, 'probability': 0.9999840260078514}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 252, 'end': 256, 'probability': 0.9999730588745024}], '被害人责任认定': [{'text': '不承担事故责任', 'start': 274, 'end': 281, 'probability': 0.9999500517702131}], '事故发生地': [{'text': '省道206线', 'start': 73, 'end': 79, 'probability': 0.6333607936532104}, {'text': '省道206线138KM+50M处', 'start': 96, 'end': 112, 'probability': 0.9999370584823595}], '被害人交通工具': [{'text': '重型专项作业车', 'start': 171, 'end': 178, 'probability': 0.8935746098784882}]}]
34 [{'犯罪嫌疑人情况': [{'text': '无证', 'start': 28, 'end': 30, 'probability': 0.9999271634934246}], '被害人': [{'text': '陈某', 'start': 42, 'end': 44, 'probability': 0.9999307405273328}, {'text': '万某', 'start': 196, 'end': 198, 'probability': 0.9999381312994444}, {'text': '李某', 'start': 220, 'end': 222, 'probability': 0.9997940166814345}, {'text': '李某', 'start': 45, 'end': 47, 'probability': 0.9999325286461271}, {'text': '万某', 'start': 223, 'end': 225, 'probability': 0.9998995089294169}, {'text': '陈某', 'start': 186, 'end': 188, 'probability': 0.9999710323464512}, {'text': '万某', 'start': 280, 'end': 282, 'probability': 0.9998394315328767}, {'text': '陈某', 'start': 274, 'end': 276, 'probability': 0.9886480636986761}, {'text': '李某', 'start': 193, 'end': 195, 'probability': 0.9999542241470749}, {'text': '李某', 'start': 277, 'end': 279, 'probability': 0.9995615886159612}, {'text': '万某', 'start': 48, 'end': 50, 'probability': 0.999928237290078}], '犯罪嫌疑人交通工具': [{'text': '正三轮摩托车', 'start': 34, 'end': 40, 'probability': 0.999976277473138}], '犯罪嫌疑人交通工具情况': [{'text': '无牌', 'start': 32, 'end': 34, 'probability': 0.9999638797872024}], '犯罪嫌疑人责任认定': [{'text': '主要责任', 'start': 252, 'end': 256, 'probability': 0.9999619725889204}], '被害人责任认定': [{'text': '无责任', 'start': 287, 'end': 290, 'probability': 0.9999725819984633}], '事故发生地': [{'text': '白金公路K7+400M路段', 'start': 75, 'end': 88, 'probability': 0.9999454028578754}, {'text': '泸西县白金公路由金马镇向白水镇方向', 'start': 51, 'end': 68, 'probability': 0.9690921336675444}], '被害人交通工具': [{'text': '轻型仓栅式货车', 'start': 156, 'end': 163, 'probability': 0.9998352588214203}, {'text': '轻型仓栅式货车', 'start': 129, 'end': 136, 'probability': 0.9999544625602113}]}]
35 [{'犯罪嫌疑人情况': [{'text': '未取得机动车驾驶证', 'start': 22, 'end': 31, 'probability': 0.9999737741254648}], '被害人': [{'text': '肖某7', 'start': 198, 'end': 201, 'probability': 0.9999525552460682}, {'text': '肖某7', 'start': 103, 'end': 106, 'probability': 0.999988555937307}, {'text': '肖某7', 'start': 115, 'end': 118, 'probability': 0.9999843836396849}, {'text': '肖某7', 'start': 157, 'end': 160, 'probability': 0.9999619725927573}], '被害人类型': [{'text': '行人', 'start': 101, 'end': 103, 'probability': 0.9999756814476513}], '犯罪嫌疑人交通工具': [{'text': '轻便二轮摩托车', 'start': 36, 'end': 43, 'probability': 0.9999707939203972}], '犯罪嫌疑人交通工具情况': [{'text': '无牌照', 'start': 33, 'end': 36, 'probability': 0.9997416279434148}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 193, 'end': 197, 'probability': 0.9998508746171808}], '被害人责任认定': [{'text': '无责任', 'start': 201, 'end': 204, 'probability': 0.9999800921405324}], '事故发生地': [{'text': '威叙公路K0+400米处', 'start': 73, 'end': 85, 'probability': 0.9999549392210838}]}]
36 [{'被害人': [{'text': '李某3', 'start': 153, 'end': 156, 'probability': 0.9999864101857412}, {'text': '李某3', 'start': 220, 'end': 223, 'probability': 0.9999791384812511}, {'text': '李某1', 'start': 203, 'end': 206, 'probability': 0.9999449260576512}, {'text': '李某1', 'start': 95, 'end': 98, 'probability': 0.9999834299773909}, {'text': '李某1', 'start': 135, 'end': 138, 'probability': 0.9999850988927506}, {'text': '李某2', 'start': 216, 'end': 219, 'probability': 0.9999758006567561}, {'text': '李某2', 'start': 142, 'end': 145, 'probability': 0.9999868870183519}], '被害人类型': [{'text': '乘车人', 'start': 139, 'end': 142, 'probability': 0.9999184615984831}, {'text': '乘车人', 'start': 150, 'end': 153, 'probability': 0.9999110712144557}, {'text': '驾驶员', 'start': 132, 'end': 135, 'probability': 0.9999649527757271}], '犯罪嫌疑人交通工具': [{'text': '重型自卸货车', 'start': 31, 'end': 37, 'probability': 0.9999806881771889}], '犯罪嫌疑人责任认定': [{'text': '主要责任', 'start': 195, 'end': 199, 'probability': 0.9999744893601843}], '被害人责任认定': [{'text': '次要责任', 'start': 208, 'end': 212, 'probability': 0.9983675549180191}, {'text': '无责任', 'start': 223, 'end': 226, 'probability': 0.9999819994667405}], '事故发生地': [{'text': '蒙自市陈家寨村路口处', 'start': 62, 'end': 72, 'probability': 0.9999669792921964}], '被害人交通工具': [{'text': '电动自行车', 'start': 106, 'end': 111, 'probability': 0.9999647142306003}]}]
37 [{'犯罪嫌疑人情况': [{'text': '无证', 'start': 33, 'end': 35, 'probability': 0.9999475482181879}], '被害人': [{'text': '李某甲', 'start': 150, 'end': 153, 'probability': 0.9998384777393312}, {'text': '李某甲', 'start': 90, 'end': 93, 'probability': 0.9999710323494355}, {'text': '李某甲', 'start': 119, 'end': 122, 'probability': 0.9999818802689191}], '犯罪嫌疑人交通工具': [{'text': '重型厢式货车', 'start': 45, 'end': 51, 'probability': 0.9999803305626642}], '犯罪嫌疑人责任认定': [{'text': '主要责任', 'start': 145, 'end': 149, 'probability': 0.9999835491825024}], '被害人责任认定': [{'text': '次要责任', 'start': 158, 'end': 162, 'probability': 0.9999740125209797}], '事故发生地': [{'text': '开通镇“某公司”门前处', 'start': 65, 'end': 76, 'probability': 0.9997841234733471}, {'text': '长白街', 'start': 80, 'end': 83, 'probability': 0.998429622436543}], '被害人交通工具': [{'text': '两轮摩托车', 'start': 104, 'end': 109, 'probability': 0.9999821186648887}]}]
38 [{'被害人': [{'text': '张某某', 'start': 97, 'end': 100, 'probability': 0.9999861717626999}, {'text': '田某某', 'start': 162, 'end': 165, 'probability': 0.9999861717661105}, {'text': '田某某', 'start': 101, 'end': 104, 'probability': 0.9999896288164791}], '被害人类型': [{'text': '乘客', 'start': 95, 'end': 97, 'probability': 0.9999518399704783}], '犯罪嫌疑人交通工具': [{'text': '小型轿车', 'start': 46, 'end': 50, 'probability': 0.9999860525617521}], '事故发生地': [{'text': '西宁殡仪馆门前', 'start': 62, 'end': 69, 'probability': 0.9999312171989061}]}]
39 [{'被害人': [{'text': '王某', 'start': 115, 'end': 117, 'probability': 0.9999624493793249}, {'text': '牛某', 'start': 118, 'end': 120, 'probability': 0.9999700786877952}], '被害人类型': [{'text': '乘车人', 'start': 112, 'end': 115, 'probability': 0.9999399190834879}], '犯罪嫌疑人交通工具': [{'text': '小型轿车', 'start': 36, 'end': 40, 'probability': 0.9999809265971038}], '犯罪嫌疑人责任认定': [{'text': '主要责任', 'start': 149, 'end': 153, 'probability': 0.9999805689709547}], '事故发生地': [{'text': '山东省茌平县信发路', 'start': 41, 'end': 50, 'probability': 0.6969953263895263}, {'text': '信发路(北环路)154号路灯杆处', 'start': 59, 'end': 75, 'probability': 0.9994540436446613}], '被害人交通工具': [{'text': '重型半挂车', 'start': 98, 'end': 103, 'probability': 0.9998182095833101}]}]
40 [{'被害人': [{'text': '张某某', 'start': 78, 'end': 81, 'probability': 0.9999898672311929}, {'text': '张某某', 'start': 85, 'end': 88, 'probability': 0.9999881983129342}, {'text': '张某某', 'start': 106, 'end': 109, 'probability': 0.9999730588716602}], '被害人类型': [{'text': '行人', 'start': 76, 'end': 78, 'probability': 0.9999779464000369}], '犯罪嫌疑人交通工具': [{'text': '小型轿车', 'start': 35, 'end': 39, 'probability': 0.9999767543112625}], '犯罪嫌疑人责任认定': [{'text': '主要责任', 'start': 141, 'end': 145, 'probability': 0.999980449763882}], '事故发生地': [{'text': '10KV辛庄线13号线杆处', 'start': 53, 'end': 66, 'probability': 0.6542351222266234}]}]
41 [{'被害人': [{'text': '张某1', 'start': 254, 'end': 257, 'probability': 0.9998345441786114}, {'text': '张某2', 'start': 177, 'end': 180, 'probability': 0.9999639991153799}, {'text': '张某2', 'start': 50, 'end': 53, 'probability': 0.9999803305639148}, {'text': '向某', 'start': 138, 'end': 140, 'probability': 0.9999748469942062}, {'text': '张某1', 'start': 173, 'end': 176, 'probability': 0.9990664950072983}, {'text': '陈某', 'start': 234, 'end': 236, 'probability': 0.9999427802838312}, {'text': '向某', 'start': 181, 'end': 183, 'probability': 0.9999516016104195}, {'text': '陈某', 'start': 43, 'end': 45, 'probability': 0.9999688866161591}, {'text': '陈某', 'start': 163, 'end': 165, 'probability': 0.9967268602576596}, {'text': '张某1', 'start': 46, 'end': 49, 'probability': 0.9998871105017031}, {'text': '刘某2', 'start': 39, 'end': 42, 'probability': 0.999957085114147}, {'text': '刘某2', 'start': 210, 'end': 213, 'probability': 0.9996982425603917}], '犯罪嫌疑人交通工具': [{'text': '小型普通客车', 'start': 28, 'end': 34, 'probability': 0.9999856949343666}], '犯罪嫌疑人责任认定': [{'text': '主要责任', 'start': 310, 'end': 314, 'probability': 0.9999479058436691}], '事故发生地': [{'text': '本区瓯江口经四路与滨水北路交叉路口', 'start': 97, 'end': 114, 'probability': 0.8203850476711239}], '被害人交通工具': [{'text': '小型轿车', 'start': 145, 'end': 149, 'probability': 0.9999843836299362}]}]
42 [{'被害人': [{'text': '王某', 'start': 280, 'end': 282, 'probability': 0.9999010583595691}, {'text': '钟某', 'start': 235, 'end': 237, 'probability': 0.9999142903414793}, {'text': '梁某', 'start': 283, 'end': 285, 'probability': 0.9996728564284112}, {'text': '郝某', 'start': 220, 'end': 222, 'probability': 0.9999645951542959}, {'text': '郝某', 'start': 228, 'end': 230, 'probability': 0.9999312173958401}, {'text': '钟某', 'start': 297, 'end': 299, 'probability': 0.9999200121071681}, {'text': '郝某', 'start': 294, 'end': 296, 'probability': 0.9999203697429238}, {'text': '钟某', 'start': 223, 'end': 225, 'probability': 0.9999700786895573}, {'text': '王某', 'start': 115, 'end': 117, 'probability': 0.9999482638238533}], '犯罪嫌疑人交通工具': [{'text': '重型仓栅式半挂车', 'start': 68, 'end': 76, 'probability': 0.9982852239953388}, {'text': '重型半挂牵引车', 'start': 44, 'end': 51, 'probability': 0.9999717475960495}], '犯罪嫌疑人责任认定': [{'text': '主要责任', 'start': 275, 'end': 279, 'probability': 0.9995352566650872}], '被害人责任认定': [{'text': '次要责任', 'start': 289, 'end': 293, 'probability': 0.9737204416069325}, {'text': '不承担事故责任', 'start': 299, 'end': 306, 'probability': 0.9998738792609174}], '事故发生地': [{'text': '91.1公里处', 'start': 97, 'end': 104, 'probability': 0.9999401577090339}, {'text': '天津市北辰区京津塘高速公路', 'start': 77, 'end': 90, 'probability': 0.9999557737594102}], '被害人交通工具': [{'text': '重型半挂牵引车', 'start': 136, 'end': 143, 'probability': 0.999632093645225}]}]
43 [{'犯罪嫌疑人情况': [{'text': '酒后', 'start': 26, 'end': 28, 'probability': 0.9999822378913166}], '被害人': [{'text': '杨某2', 'start': 97, 'end': 100, 'probability': 0.9999872446448137}, {'text': '张某某', 'start': 116, 'end': 119, 'probability': 0.9999812842198708}, {'text': '张某某', 'start': 161, 'end': 164, 'probability': 0.9999721051793387}, {'text': '杨某1', 'start': 169, 'end': 172, 'probability': 0.9999476677980397}, {'text': '杨某2', 'start': 157, 'end': 160, 'probability': 0.9999794961066044}, {'text': '杨某1', 'start': 136, 'end': 139, 'probability': 0.999978065606399}], '被害人类型': [{'text': '行人', 'start': 95, 'end': 97, 'probability': 0.9999821186841018}], '犯罪嫌疑人交通工具': [{'text': '小型越野车', 'start': 38, 'end': 43, 'probability': 0.9999833107687408}], '事故发生地': [{'text': '襄城区卧龙镇胡巷村6组路段', 'start': 66, 'end': 79, 'probability': 0.9999785424422925}]}]
44 [{'犯罪嫌疑人情况': [{'text': '违法操作', 'start': 73, 'end': 77, 'probability': 0.9283558174492441}], '被害人': [{'text': '杨某1', 'start': 108, 'end': 111, 'probability': 0.9999843836396849}], '犯罪嫌疑人交通工具': [{'text': '轿车', 'start': 36, 'end': 38, 'probability': 0.9999483829298867}], '犯罪嫌疑人责任认定': [{'text': '主要责任', 'start': 196, 'end': 200, 'probability': 0.9999551777138436}], '事故发生地': [{'text': '临清市先锋路街道办事处石槽村南路口20米处', 'start': 49, 'end': 70, 'probability': 0.9999816418535374}]}]
45 [{'被害人': [{'text': '闫某1', 'start': 94, 'end': 97, 'probability': 0.999968052163851}, {'text': '刘某', 'start': 79, 'end': 81, 'probability': 0.9999824763101799}, {'text': '刘某', 'start': 120, 'end': 122, 'probability': 0.9999742509571661}, {'text': '闫某1', 'start': 123, 'end': 126, 'probability': 0.9999822378887586}], '犯罪嫌疑人交通工具': [{'text': '重型特殊结构货车', 'start': 33, 'end': 41, 'probability': 0.9999743701591228}], '被害人交通工具情况': [{'text': '无号牌', 'start': 83, 'end': 86, 'probability': 0.9999665024652984}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 159, 'end': 163, 'probability': 0.9999809265732296}], '事故发生地': [{'text': '虞某某大道与解放东路交叉路口', 'start': 56, 'end': 70, 'probability': 0.9999414688855381}], '被害人交通工具': [{'text': '二轮电动车', 'start': 86, 'end': 91, 'probability': 0.9999872446367135}]}]
46 [{'被害人': [{'text': '王某', 'start': 96, 'end': 98, 'probability': 0.9999767543228302}, {'text': '王某', 'start': 145, 'end': 147, 'probability': 0.9999787808590241}, {'text': '王某', 'start': 106, 'end': 108, 'probability': 0.9999818802698428}], '犯罪嫌疑人交通工具': [{'text': '全挂大货车', 'start': 35, 'end': 40, 'probability': 0.9781577207045729}], '犯罪嫌疑人责任认定': [{'text': '主要责任', 'start': 187, 'end': 191, 'probability': 0.9999274024568194}], '事故发生地': [{'text': '准格尔旗曹羊线35km+300m处', 'start': 47, 'end': 64, 'probability': 0.9999588732173663}], '被害人交通工具': [{'text': '小型普通客车', 'start': 82, 'end': 88, 'probability': 0.9999827147217388}]}]
47 [{'犯罪嫌疑人情况': [{'text': '未依法取得机动车驾驶证', 'start': 109, 'end': 120, 'probability': 0.9604953777573328}], '被害人': [{'text': '罗某', 'start': 59, 'end': 61, 'probability': 0.9999780656086017}, {'text': '罗某', 'start': 186, 'end': 188, 'probability': 0.9994675629435825}], '犯罪嫌疑人交通工具': [{'text': '轻型普通货车', 'start': 34, 'end': 40, 'probability': 0.9999802113512288}], '事故发生地': [{'text': '209国道3093KM+650M(某公司前路段)', 'start': 77, 'end': 101, 'probability': 0.9949218325845166}], '被害人交通工具': [{'text': '自行车', 'start': 62, 'end': 65, 'probability': 0.9999805689801633}]}]
48 [{'被害人': [{'text': '闫某某', 'start': 85, 'end': 88, 'probability': 0.9999911785319711}, {'text': '闫某某', 'start': 77, 'end': 80, 'probability': 0.9999921322003189}], '被害人类型': [{'text': '行人', 'start': 75, 'end': 77, 'probability': 0.9999800921443125}], '犯罪嫌疑人交通工具': [{'text': '中型厢货车', 'start': 38, 'end': 43, 'probability': 0.9999735356900743}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 135, 'end': 139, 'probability': 0.9999834299563304}], '事故发生地': [{'text': '该村中心十字路口', 'start': 62, 'end': 70, 'probability': 0.8160873841865168}]}]
49 [{'被害人': [{'text': '王某', 'start': 62, 'end': 64, 'probability': 0.9999830723516112}], '犯罪嫌疑人交通工具': [{'text': '五菱面包车', 'start': 31, 'end': 36, 'probability': 0.9999374157269472}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 85, 'end': 89, 'probability': 0.9999465946923323}], '事故发生地': [{'text': '费县梁邱镇营子村路段', 'start': 40, 'end': 50, 'probability': 0.9999758006606214}]}]
50 [{'被害人': [{'text': '李某2', 'start': 77, 'end': 80, 'probability': 0.9999893904003443}, {'text': '李某2', 'start': 170, 'end': 173, 'probability': 0.999962568630167}, {'text': '李某2', 'start': 223, 'end': 226, 'probability': 0.9999717475935768}, {'text': '李某2', 'start': 404, 'end': 407, 'probability': 0.9999499327245758}, {'text': '李某2', 'start': 186, 'end': 189, 'probability': 0.999985694932775}], '犯罪嫌疑人交通工具': [{'text': '重型半挂牵引车', 'start': 32, 'end': 39, 'probability': 0.9993390523621386}, {'text': '挂车', 'start': 47, 'end': 49, 'probability': 0.9996306630946492}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 264, 'end': 268, 'probability': 0.9999871254349699}], '事故发生地': [{'text': '省道215线16KM+300M处', 'start': 120, 'end': 136, 'probability': 0.9999507670655134}], '被害人交通工具': [{'text': '小型普通客车', 'start': 85, 'end': 91, 'probability': 0.9999817610610648}]}]
51 [{'被害人': [{'text': '孙某某', 'start': 66, 'end': 69, 'probability': 0.9999886751473355}, {'text': '唐某某', 'start': 84, 'end': 87, 'probability': 0.9999748469974747}, {'text': '孙某', 'start': 79, 'end': 81, 'probability': 0.999972582051214}, {'text': '唐某某', 'start': 62, 'end': 65, 'probability': 0.9999872446466895}], '被害人类型': [{'text': '行人', 'start': 60, 'end': 62, 'probability': 0.9999755622441455}], '犯罪嫌疑人交通工具': [{'text': '小型轿车', 'start': 32, 'end': 36, 'probability': 0.9999729396619728}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 113, 'end': 117, 'probability': 0.9999626876605703}], '事故发生地': [{'text': '滦县陈庄村东处', 'start': 51, 'end': 58, 'probability': 0.9997022791730927}]}]
52 [{'被害人': [{'text': '张某某1', 'start': 107, 'end': 111, 'probability': 0.9096534969684811}, {'text': '林某某1', 'start': 255, 'end': 259, 'probability': 0.9039757704830791}, {'text': '林某某1', 'start': 215, 'end': 219, 'probability': 0.9999655488066423}, {'text': '林某某1', 'start': 68, 'end': 72, 'probability': 0.9999850988943138}], '被害人类型': [{'text': '行人', 'start': 66, 'end': 68, 'probability': 0.999983191560375}], '犯罪嫌疑人交通工具': [{'text': '轿车', 'start': 30, 'end': 32, 'probability': 0.8971955318590119}], '犯罪嫌疑人责任认定': [{'text': '主要责任', 'start': 246, 'end': 250, 'probability': 0.9999492173773348}], '被害人责任认定': [{'text': '次要责任', 'start': 266, 'end': 270, 'probability': 0.8237402897961488}], '事故发生地': [{'text': '大肇路45公里90米处', 'start': 46, 'end': 57, 'probability': 0.99986565441057}]}]
53 [{'犯罪嫌疑人情况': [{'text': '酒后', 'start': 28, 'end': 30, 'probability': 0.9996048836301199}, {'text': '无证', 'start': 30, 'end': 32, 'probability': 0.9999666216419314}], '被害人': [{'text': '凌某', 'start': 104, 'end': 106, 'probability': 0.9999768735160615}, {'text': '凌某', 'start': 114, 'end': 116, 'probability': 0.9999691250101534}, {'text': '凌某', 'start': 159, 'end': 161, 'probability': 0.9999514823409754}], '被害人类型': [{'text': '行人', 'start': 102, 'end': 104, 'probability': 0.9999641183078154}, {'text': '车上人员', 'start': 133, 'end': 137, 'probability': 0.9900302328948527}], '犯罪嫌疑人交通工具': [{'text': '两轮摩托车', 'start': 37, 'end': 42, 'probability': 0.9999843836386333}], '犯罪嫌疑人交通工具情况': [{'text': '无号牌', 'start': 34, 'end': 37, 'probability': 0.9999763966748816}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 206, 'end': 210, 'probability': 0.9999607804066954}], '事故发生地': [{'text': '雁江区资资路39KM+350M处', 'start': 76, 'end': 92, 'probability': 0.999958038628165}]}]
54 [{'被害人': [{'text': '孙某某', 'start': 191, 'end': 194, 'probability': 0.9999597076523514}, {'text': '刘某某', 'start': 195, 'end': 198, 'probability': 0.9999549393620555}, {'text': '魏某某', 'start': 86, 'end': 89, 'probability': 0.9997835269142428}, {'text': '马某某', 'start': 182, 'end': 185, 'probability': 0.99996995946303}, {'text': '魏某某', 'start': 226, 'end': 229, 'probability': 0.9896747832878106}, {'text': '杨某某', 'start': 178, 'end': 181, 'probability': 0.99997067472016}, {'text': '魏某某', 'start': 171, 'end': 174, 'probability': 0.9999824763093272}], '被害人类型': [{'text': '驾驶员', 'start': 168, 'end': 171, 'probability': 0.9997531873442043}, {'text': '乘车人', 'start': 175, 'end': 178, 'probability': 0.354587265926142}], '犯罪嫌疑人交通工具': [{'text': '半挂车', 'start': 31, 'end': 34, 'probability': 0.9999790192736668}], '犯罪嫌疑人责任认定': [{'text': '主要责任', 'start': 221, 'end': 225, 'probability': 0.9999625686292575}], '被害人责任认定': [{'text': '次要责任', 'start': 235, 'end': 239, 'probability': 0.9649214774548369}], '事故发生地': [{'text': '亳州市谯城区', 'start': 35, 'end': 41, 'probability': 0.9888578286889498}, {'text': '下行线494KM+50M处', 'start': 53, 'end': 66, 'probability': 0.9997631451036284}], '被害人交通工具': [{'text': '小轿车', 'start': 102, 'end': 105, 'probability': 0.9999738933359765}, {'text': '小轿车', 'start': 165, 'end': 168, 'probability': 0.999895336679586}]}]
55 [{'被害人': [{'text': '杨某某', 'start': 108, 'end': 111, 'probability': 0.999988675141708}, {'text': '杨某某', 'start': 187, 'end': 190, 'probability': 0.9999769927137834}, {'text': '杨某某', 'start': 77, 'end': 80, 'probability': 0.999987125435382}, {'text': '杨某某', 'start': 121, 'end': 124, 'probability': 0.9999880791045257}, {'text': '杨某某', 'start': 181, 'end': 184, 'probability': 0.9999800921353597}], '犯罪嫌疑人交通工具': [{'text': '小客车', 'start': 30, 'end': 33, 'probability': 0.9999673368951107}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 235, 'end': 239, 'probability': 0.9999769926987767}], '事故发生地': [{'text': '莞樟路黄江镇社贝天桥路段', 'start': 52, 'end': 64, 'probability': 0.9999634030291986}], '被害人交通工具': [{'text': '自行车', 'start': 82, 'end': 85, 'probability': 0.9999760390647907}, {'text': '摩托车', 'start': 95, 'end': 98, 'probability': 0.9999512439375735}]}]
56 [{'被害人': [{'text': '荫全志', 'start': 98, 'end': 101, 'probability': 0.9999862909782564}, {'text': '荫全志', 'start': 119, 'end': 122, 'probability': 0.9999706747283454}, {'text': '荫全志', 'start': 157, 'end': 160, 'probability': 0.9999814034352426}], '犯罪嫌疑人交通工具': [{'text': '小型轿车', 'start': 30, 'end': 34, 'probability': 0.9999386078614236}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 152, 'end': 156, 'probability': 0.7355406616038422}], '被害人责任认定': [{'text': '无责任', 'start': 160, 'end': 163, 'probability': 0.9999862909694457}], '事故发生地': [{'text': '66公里900米(固安县郭翟村红绿灯南侧)', 'start': 54, 'end': 75, 'probability': 0.995525191400958}], '被害人交通工具': [{'text': '电动三轮车', 'start': 104, 'end': 109, 'probability': 0.9999849796631821}]}]
57 [{'被害人': [{'text': '夏某1', 'start': 125, 'end': 128, 'probability': 0.9999831915589255}, {'text': '夏某1', 'start': 102, 'end': 105, 'probability': 0.9999874830627391}], '犯罪嫌疑人交通工具': [{'text': '重型仓栅式货车', 'start': 34, 'end': 41, 'probability': 0.99998044976887}], '犯罪嫌疑人责任认定': [{'text': '主要责任', 'start': 170, 'end': 174, 'probability': 0.9999797345114985}], '被害人责任认定': [{'text': '次要责任', 'start': 200, 'end': 204, 'probability': 0.601288905604271}], '事故发生地': [{'text': '句容市区宝华山路与句卓路十字交叉路口处', 'start': 62, 'end': 81, 'probability': 0.9999159591818341}], '被害人交通工具': [{'text': '电动自行车', 'start': 107, 'end': 112, 'probability': 0.9999874830505462}]}]
58 [{'被害人': [{'text': '李某1', 'start': 131, 'end': 134, 'probability': 0.9999756814486744}, {'text': '李某1', 'start': 78, 'end': 81, 'probability': 0.9999895096072464}, {'text': '李某1', 'start': 144, 'end': 147, 'probability': 0.9998474155761272}], '犯罪嫌疑人交通工具': [{'text': '电动三轮摩托车', 'start': 30, 'end': 37, 'probability': 0.9999773503357261}], '犯罪嫌疑人交通工具情况': [{'text': '无号牌', 'start': 27, 'end': 30, 'probability': 0.9999819994641825}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 126, 'end': 130, 'probability': 0.9999812841996913}], '被害人责任认定': [{'text': '无责任', 'start': 134, 'end': 137, 'probability': 0.9999870062252967}], '事故发生地': [{'text': '济川一小学校外路段', 'start': 63, 'end': 72, 'probability': 0.9999747277791471}]}]
59 [{'被害人': [{'text': '王某', 'start': 137, 'end': 139, 'probability': 0.9988350863296063}, {'text': '王某', 'start': 93, 'end': 95, 'probability': 0.9999741317508608}, {'text': '关某', 'start': 126, 'end': 128, 'probability': 0.9999496943021313}, {'text': '王某', 'start': 112, 'end': 114, 'probability': 0.9999840260189359}], '被害人类型': [{'text': '乘车人', 'start': 123, 'end': 126, 'probability': 0.9997628420175602}], '犯罪嫌疑人交通工具': [{'text': '重型仓栅式半挂车', 'start': 48, 'end': 56, 'probability': 0.999973058856952}, {'text': '重型半挂牵引车', 'start': 34, 'end': 41, 'probability': 0.999972462840276}], '被害人交通工具情况': [{'text': '无号牌', 'start': 98, 'end': 101, 'probability': 0.999948144606833}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 173, 'end': 177, 'probability': 0.9999794960688746}], '事故发生地': [{'text': 'G237线392KM+350M处', 'start': 64, 'end': 80, 'probability': 0.9999735357032051}], '被害人交通工具': [{'text': '三轮电动车', 'start': 101, 'end': 106, 'probability': 0.9999856949190047}]}]
60 [{'被害人': [{'text': '罗某', 'start': 85, 'end': 87, 'probability': 0.9997944937140346}, {'text': '张某', 'start': 62, 'end': 64, 'probability': 0.9999700786797803}, {'text': '张某', 'start': 42, 'end': 44, 'probability': 0.9999455220953735}]}]
61 [{'被害人': [{'text': '李某1', 'start': 30, 'end': 33, 'probability': 0.999835852935405}]}]
62 [{'犯罪嫌疑人情况': [{'text': '酒后', 'start': 28, 'end': 30, 'probability': 0.9999339584931732}, {'text': '无证', 'start': 30, 'end': 32, 'probability': 0.9999539856503361}], '被害人': [{'text': '方某', 'start': 221, 'end': 223, 'probability': 0.9999711515383751}, {'text': '方某', 'start': 133, 'end': 135, 'probability': 0.9999793768803045}], '犯罪嫌疑人交通工具': [{'text': '三轮机动车', 'start': 49, 'end': 54, 'probability': 0.9999533896070432}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 213, 'end': 217, 'probability': 0.9999772311308277}], '被害人责任认定': [{'text': '不负事故责任', 'start': 223, 'end': 229, 'probability': 0.9998720896257396}], '事故发生地': [{'text': '宣城市宣州区杨华线由北向南方向', 'start': 56, 'end': 71, 'probability': 0.9979029658913277}, {'text': '华阳社区街道“华阳男女浴室”门前路段', 'start': 77, 'end': 95, 'probability': 0.9999393230121711}], '被害人交通工具': [{'text': '三轮载货摩托车', 'start': 114, 'end': 121, 'probability': 0.6779959354995597}]}]
63 [{'犯罪嫌疑人情况': [{'text': '未取得机动车驾驶证', 'start': 25, 'end': 34, 'probability': 0.9999843836424702}], '被害人': [{'text': '陈某', 'start': 119, 'end': 121, 'probability': 0.9999752046186927}, {'text': '陈某', 'start': 129, 'end': 131, 'probability': 0.9999819994690426}, {'text': '陈某', 'start': 190, 'end': 192, 'probability': 0.9999520782976248}], '被害人类型': [{'text': '行人', 'start': 117, 'end': 119, 'probability': 0.9999785424426335}], '犯罪嫌疑人交通工具': [{'text': '电动自行车', 'start': 46, 'end': 51, 'probability': 0.9999852181010027}], '犯罪嫌疑人交通工具情况': [{'text': '不予登记', 'start': 38, 'end': 42, 'probability': 0.9991898088745401}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 226, 'end': 230, 'probability': 0.9999691249425098}], '事故发生地': [{'text': '环山乡中埠大桥', 'start': 88, 'end': 95, 'probability': 0.9999526744342973}]}]
64 [{'被害人': [{'text': '李4', 'start': 101, 'end': 103, 'probability': 0.9999132175055365}, {'text': '李4', 'start': 212, 'end': 214, 'probability': 0.9992880112118243}, {'text': '李4', 'start': 84, 'end': 86, 'probability': 0.9999760390678034}, {'text': '李4', 'start': 169, 'end': 171, 'probability': 0.9998930720909982}], '被害人类型': [{'text': '乘车人', 'start': 133, 'end': 136, 'probability': 0.5616246922476087}], '犯罪嫌疑人交通工具': [{'text': '重型半挂牵引车', 'start': 35, 'end': 42, 'probability': 0.9992625199539091}], '事故发生地': [{'text': '后石村天源加气站附近路口', 'start': 61, 'end': 73, 'probability': 0.9999722244279781}, {'text': '离石区信义工业大道', 'start': 45, 'end': 54, 'probability': 0.9932806680629653}], '被害人交通工具': [{'text': '电动二轮车', 'start': 92, 'end': 97, 'probability': 0.9999860525458075}]}]
65 [{'犯罪嫌疑人情况': [{'text': '违规载人', 'start': 79, 'end': 83, 'probability': 0.6962519281543678}], '被害人': [{'text': '乔某2', 'start': 170, 'end': 173, 'probability': 0.9996478191028757}, {'text': '乔某2', 'start': 105, 'end': 108, 'probability': 0.9999885559397939}, {'text': '乔某2', 'start': 117, 'end': 120, 'probability': 0.9999406345891799}, {'text': '乔某2', 'start': 129, 'end': 132, 'probability': 0.9999549392758951}], '犯罪嫌疑人交通工具': [{'text': '轻型仓栅式货车', 'start': 32, 'end': 39, 'probability': 0.9999861717693079}], '事故发生地': [{'text': '张店镇龚庄路段', 'start': 53, 'end': 60, 'probability': 0.9999181048050332}]}]
66 [{'被害人': [{'text': '许某2', 'start': 70, 'end': 73, 'probability': 0.9999885559402486}, {'text': '许某2', 'start': 148, 'end': 151, 'probability': 0.9999850988927506}, {'text': '许某2', 'start': 86, 'end': 89, 'probability': 0.9999889135645788}], '犯罪嫌疑人交通工具': [{'text': '小型轿车', 'start': 32, 'end': 36, 'probability': 0.9999815226439068}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 189, 'end': 193, 'probability': 0.9999774695334906}], '事故发生地': [{'text': 'S254省道', 'start': 38, 'end': 44, 'probability': 0.9995574463383861}, {'text': '45K+750m处', 'start': 51, 'end': 60, 'probability': 0.9998675617780464}], '被害人交通工具': [{'text': '电动三轮车', 'start': 76, 'end': 81, 'probability': 0.9999852180862945}]}]
67 [{'被害人': [{'text': '付某', 'start': 95, 'end': 97, 'probability': 0.9999595883503218}, {'text': '付某', 'start': 130, 'end': 132, 'probability': 0.9999657872180023}, {'text': '付某', 'start': 76, 'end': 78, 'probability': 0.9999791384631465}], '犯罪嫌疑人交通工具': [{'text': '小型轿车', 'start': 30, 'end': 34, 'probability': 0.9999771119445882}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 122, 'end': 126, 'probability': 0.9999868870115307}], '被害人责任认定': [{'text': '无事故责任', 'start': 132, 'end': 137, 'probability': 0.9999700786000574}], '事故发生地': [{'text': '沈阳经济技术开发区', 'start': 48, 'end': 57, 'probability': 0.9999828339357464}], '被害人交通工具': [{'text': '两轮电动车', 'start': 67, 'end': 72, 'probability': 0.999967575176413}]}]
68 [{'被害人': [{'text': '何某', 'start': 29, 'end': 31, 'probability': 0.9999814034336509}]}]
69 [{'被害人': [{'text': '李XX', 'start': 95, 'end': 98, 'probability': 0.9999734164782126}, {'text': '李XX', 'start': 214, 'end': 217, 'probability': 0.9908055691666959}, {'text': '李XX', 'start': 112, 'end': 115, 'probability': 0.9999663831952148}, {'text': '李XX', 'start': 173, 'end': 176, 'probability': 0.9999512439759428}, {'text': '李XX', 'start': 251, 'end': 254, 'probability': 0.9993494615319065}, {'text': '李XX', 'start': 294, 'end': 297, 'probability': 0.9996318474790797}, {'text': '李XX', 'start': 82, 'end': 85, 'probability': 0.9999842644315748}], '被害人类型': [{'text': '行人', 'start': 80, 'end': 82, 'probability': 0.9999680521604404}], '犯罪嫌疑人交通工具': [{'text': '小型普通客车', 'start': 41, 'end': 47, 'probability': 0.9999859333512546}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 154, 'end': 158, 'probability': 0.9999854565132864}], '事故发生地': [{'text': '滨河市场前交叉路口', 'start': 66, 'end': 75, 'probability': 0.9999618533684043}]}]
70 [{'犯罪嫌疑人情况': [{'text': '酒后', 'start': 31, 'end': 33, 'probability': 0.9999721051808592}], '被害人': [{'text': '李某6', 'start': 216, 'end': 219, 'probability': 0.9999333631101024}, {'text': '李某6', 'start': 129, 'end': 132, 'probability': 0.9999803305638579}, {'text': '李某6', 'start': 98, 'end': 101, 'probability': 0.9999893903994916}, {'text': '李某6', 'start': 88, 'end': 91, 'probability': 0.9999883175201063}], '犯罪嫌疑人交通工具': [{'text': '面包车', 'start': 52, 'end': 55, 'probability': 0.9999783040158405}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 211, 'end': 215, 'probability': 0.9999246607276291}], '被害人责任认定': [{'text': '无责任', 'start': 219, 'end': 222, 'probability': 0.999987483060778}], '事故发生地': [{'text': '辉县市高庄乡中太石化加油站附近', 'start': 67, 'end': 82, 'probability': 0.9999775887770852}]}]
71 [{'犯罪嫌疑人情况': [{'text': '酒后', 'start': 30, 'end': 32, 'probability': 0.9999748469558654}], '被害人': [{'text': '瞿某1', 'start': 103, 'end': 106, 'probability': 0.9999802113342184}, {'text': '瞿某1', 'start': 129, 'end': 132, 'probability': 0.9999864101839648}, {'text': '瞿某1', 'start': 86, 'end': 89, 'probability': 0.9999846220592019}], '犯罪嫌疑人交通工具': [{'text': '小型轿车', 'start': 37, 'end': 41, 'probability': 0.9999765159059564}], '犯罪嫌疑人责任认定': [{'text': '主要责任', 'start': 244, 'end': 248, 'probability': 0.9999667408540347}], '事故发生地': [{'text': '启东市海复镇均里中心路交叉路口', 'start': 58, 'end': 73, 'probability': 0.9999538665063028}], '被害人交通工具': [{'text': '三轮电动车', 'start': 92, 'end': 97, 'probability': 0.9999885559327595}]}]
72 [{'被害人': [{'text': '李某1', 'start': 41, 'end': 44, 'probability': 0.9991269320827492}, {'text': '苏某', 'start': 142, 'end': 144, 'probability': 0.9998736418419867}, {'text': '彭某', 'start': 45, 'end': 47, 'probability': 0.986460157930587}, {'text': '李某1', 'start': 159, 'end': 162, 'probability': 0.9998532583729514}, {'text': '陈某', 'start': 38, 'end': 40, 'probability': 0.9998368064704266}, {'text': '陈某', 'start': 156, 'end': 158, 'probability': 0.9999457604876909}, {'text': '吴某', 'start': 35, 'end': 37, 'probability': 0.9998062907260419}, {'text': '吴某', 'start': 231, 'end': 233, 'probability': 0.99990248907568}, {'text': '吴某', 'start': 101, 'end': 103, 'probability': 0.990092026376626}, {'text': '吴某', 'start': 149, 'end': 151, 'probability': 0.9999433763818502}, {'text': '苏某', 'start': 32, 'end': 34, 'probability': 0.9996957352777969}], '被害人类型': [{'text': '乘员', 'start': 131, 'end': 133, 'probability': 0.9973944498986782}], '犯罪嫌疑人交通工具': [{'text': '小型普通客车', 'start': 54, 'end': 60, 'probability': 0.9998465826443237}], '犯罪嫌疑人责任认定': [{'text': '主要责任', 'start': 226, 'end': 230, 'probability': 0.9999816418467873}], '被害人责任认定': [{'text': '次要责任', 'start': 237, 'end': 241, 'probability': 0.999973297244523}], '事故发生地': [{'text': '沙厦高速公路厦门往沙县方向', 'start': 62, 'end': 75, 'probability': 0.992125815713365}, {'text': '沙厦高速(下行)186公里470米处', 'start': 80, 'end': 98, 'probability': 0.9999668600736555}], '被害人交通工具': [{'text': '轻型普通货车', 'start': 109, 'end': 115, 'probability': 0.999940038512392}]}]
73 [{'犯罪嫌疑人情况': [{'text': '酒后', 'start': 30, 'end': 32, 'probability': 0.9999771119384917}], '被害人': [{'text': '杨某1', 'start': 96, 'end': 99, 'probability': 0.9999850988932906}, {'text': '杨某1', 'start': 168, 'end': 171, 'probability': 0.9999859333464229}, {'text': '杨某1', 'start': 233, 'end': 236, 'probability': 0.9999744893573279}, {'text': '杨某1', 'start': 114, 'end': 117, 'probability': 0.9999859333501036}], '犯罪嫌疑人交通工具': [{'text': '小型轿车', 'start': 40, 'end': 44, 'probability': 0.9999817610610648}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 225, 'end': 229, 'probability': 0.999982118665983}], '被害人责任认定': [{'text': '不负事故责任', 'start': 236, 'end': 242, 'probability': 0.9999754429941277}], '事故发生地': [{'text': '惠州市惠某区镇隆某X204线14km+300m路段', 'start': 66, 'end': 91, 'probability': 0.9999682905779537}], '被害人交通工具': [{'text': '自行车', 'start': 102, 'end': 105, 'probability': 0.9999797345159323}]}]
74 [{'被害人': [{'text': '李某', 'start': 171, 'end': 173, 'probability': 0.9982302416611333}, {'text': '李某', 'start': 119, 'end': 121, 'probability': 0.9999618533875037}, {'text': '李某', 'start': 41, 'end': 43, 'probability': 0.999963164666525}], '犯罪嫌疑人交通工具': [{'text': '小型普通客车', 'start': 34, 'end': 40, 'probability': 0.9999848604773547}], '事故发生地': [{'text': '麒麟区S315线', 'start': 79, 'end': 87, 'probability': 0.9999728204502389}, {'text': 'K27+720米处', 'start': 94, 'end': 103, 'probability': 0.9999666215932024}]}]
75 [{'被害人': [{'text': '载某某', 'start': 87, 'end': 90, 'probability': 0.5616490653962387}, {'text': '张某', 'start': 76, 'end': 78, 'probability': 0.9999313365784559}, {'text': '霍某某', 'start': 102, 'end': 105, 'probability': 0.9999812842183644}, {'text': '张某', 'start': 115, 'end': 117, 'probability': 0.9999179856520186}], '犯罪嫌疑人交通工具': [{'text': '挂货车', 'start': 42, 'end': 45, 'probability': 0.9999595884591201}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 145, 'end': 149, 'probability': 0.9999793768658094}], '事故发生地': [{'text': '北岗底村北路段', 'start': 59, 'end': 66, 'probability': 0.9999070181768843}], '被害人交通工具': [{'text': '二轮摩托车', 'start': 81, 'end': 86, 'probability': 0.9999847412486247}]}]
76 [{'被害人': [{'text': '谭某某', 'start': 100, 'end': 103, 'probability': 0.9999860525574036}, {'text': '谭某某', 'start': 90, 'end': 93, 'probability': 0.9999901056504541}], '被害人类型': [{'text': '行人', 'start': 84, 'end': 86, 'probability': 0.9999419457916758}], '犯罪嫌疑人交通工具': [{'text': '普通二轮摩托车', 'start': 39, 'end': 46, 'probability': 0.999981284198384}], '犯罪嫌疑人责任认定': [{'text': '主要责任', 'start': 147, 'end': 151, 'probability': 0.9999834299763677}], '事故发生地': [{'text': '罗城街道兴华二路XXX号门前路段', 'start': 66, 'end': 82, 'probability': 0.9999718668027526}]}]
77 [{'被害人': [{'text': '李某某', 'start': 70, 'end': 73, 'probability': 0.902656673912535}, {'text': '李某某', 'start': 324, 'end': 327, 'probability': 0.9999377736607045}, {'text': '李某某', 'start': 161, 'end': 164, 'probability': 0.9999880791058757}, {'text': '赖某某', 'start': 165, 'end': 168, 'probability': 0.9999859333512546}, {'text': '赖某某', 'start': 54, 'end': 57, 'probability': 0.7647065040820848}, {'text': '赖某某', 'start': 214, 'end': 217, 'probability': 0.9998470579115235}], '被害人类型': [{'text': '后排乘客', 'start': 157, 'end': 161, 'probability': 0.9998224989885784}], '犯罪嫌疑人交通工具': [{'text': '小型越野客车', 'start': 37, 'end': 43, 'probability': 0.9999800921472968}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 515, 'end': 519, 'probability': 0.9726386378815732}], '事故发生地': [{'text': 'S11平汝高速公路', 'start': 100, 'end': 109, 'probability': 0.9987586078405286}, {'text': '本市北盛镇出口路段', 'start': 117, 'end': 126, 'probability': 0.9995203087984805}]}]
78 [{'被害人': [{'text': '郑某某', 'start': 89, 'end': 92, 'probability': 0.9999884367319112}, {'text': '郑某某', 'start': 75, 'end': 78, 'probability': 0.9999799729391157}, {'text': '郑某某', 'start': 143, 'end': 146, 'probability': 0.9999719860092853}, {'text': '郑某某', 'start': 106, 'end': 109, 'probability': 0.9999735357113906}], '犯罪嫌疑人交通工具': [{'text': '小型轿车', 'start': 35, 'end': 39, 'probability': 0.9999805689791685}], '犯罪嫌疑人责任认定': [{'text': '主要责任', 'start': 138, 'end': 142, 'probability': 0.999982833925614}], '被害人责任认定': [{'text': '次要责任', 'start': 152, 'end': 156, 'probability': 0.9999713899035072}], '事故发生地': [{'text': '与立城大道交叉口西500米路段', 'start': 56, 'end': 71, 'probability': 0.9536543251167586}, {'text': '淮滨县城关淮河大道', 'start': 40, 'end': 49, 'probability': 0.9813889463466126}], '被害人交通工具': [{'text': '两轮电动车', 'start': 81, 'end': 86, 'probability': 0.9999786616056383}]}]
79 [{'犯罪嫌疑人情况': [{'text': '酒后', 'start': 25, 'end': 27, 'probability': 0.9999772311353183}], '被害人': [{'text': '李某', 'start': 160, 'end': 162, 'probability': 0.9999016547418194}, {'text': '熊某', 'start': 208, 'end': 210, 'probability': 0.9998756684645116}, {'text': '徐某', 'start': 118, 'end': 120, 'probability': 0.999938369617297}, {'text': '熊某', 'start': 38, 'end': 40, 'probability': 0.9964847776063834}, {'text': '熊某', 'start': 157, 'end': 159, 'probability': 0.9994832288024824}, {'text': '李某', 'start': 205, 'end': 207, 'probability': 0.9998685162556171}, {'text': '徐某', 'start': 202, 'end': 204, 'probability': 0.9997954468130246}, {'text': '徐某', 'start': 44, 'end': 46, 'probability': 0.9998698270475614}, {'text': '李某', 'start': 41, 'end': 43, 'probability': 0.9997057489122128}], '犯罪嫌疑人交通工具': [{'text': '出租车', 'start': 32, 'end': 35, 'probability': 0.9999760390747667}], '犯罪嫌疑人责任认定': [{'text': '主要责任', 'start': 184, 'end': 188, 'probability': 0.9996467533752309}], '被害人责任认定': [{'text': '不承担此次事故责任', 'start': 210, 'end': 219, 'probability': 0.9999858141370055}], '事故发生地': [{'text': '桥中段', 'start': 63, 'end': 66, 'probability': 0.9834608989953182}, {'text': '武汉市江汉二桥', 'start': 49, 'end': 56, 'probability': 0.9978250607446171}], '被害人交通工具': [{'text': '轿车', 'start': 88, 'end': 90, 'probability': 0.99955658002294}]}]
80 [{'犯罪嫌疑人情况': [{'text': '未依法取得机动车驾驶证', 'start': 28, 'end': 39, 'probability': 0.9999800921475526}], '被害人': [{'text': '王某某', 'start': 115, 'end': 118, 'probability': 0.9999804497709164}, {'text': '王某某', 'start': 133, 'end': 136, 'probability': 0.9999392041549982}, {'text': '王某某', 'start': 152, 'end': 155, 'probability': 0.999191028117437}], '犯罪嫌疑人交通工具': [{'text': '出租车', 'start': 57, 'end': 60, 'probability': 0.9999742509568108}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 205, 'end': 209, 'probability': 0.9999728204028742}], '事故发生地': [{'text': '杜尔伯特蒙古族自治县泰康镇府前路', 'start': 62, 'end': 78, 'probability': 0.9999321709746596}, {'text': '林业局路口', 'start': 85, 'end': 90, 'probability': 0.9999313364559725}], '被害人交通工具': [{'text': '自行车', 'start': 121, 'end': 124, 'probability': 0.9999809265900694}]}]
81 [{'被害人': [{'text': '麻某某', 'start': 114, 'end': 117, 'probability': 0.9999862909748174}, {'text': '麻某某', 'start': 93, 'end': 96, 'probability': 0.9999842644320438}, {'text': '麻某某', 'start': 129, 'end': 132, 'probability': 0.9999836683824697}], '犯罪嫌疑人交通工具': [{'text': '装载机', 'start': 41, 'end': 44, 'probability': 0.999957323496929}], '犯罪嫌疑人交通工具情况': [{'text': '无号牌', 'start': 31, 'end': 34, 'probability': 0.9999752045815598}], '被害人交通工具情况': [{'text': '无号牌', 'start': 99, 'end': 102, 'probability': 0.9999212040767702}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 156, 'end': 160, 'probability': 0.9999789000322323}], '事故发生地': [{'text': '村中巷道', 'start': 76, 'end': 80, 'probability': 0.9929106370592606}, {'text': '察右后旗乌兰哈达苏木白某某圐圙新村西路', 'start': 46, 'end': 65, 'probability': 0.99928074480421}], '被害人交通工具': [{'text': '三轮翻斗车', 'start': 105, 'end': 110, 'probability': 0.9999797345029293}]}]
82 [{'犯罪嫌疑人情况': [{'text': '无证', 'start': 31, 'end': 33, 'probability': 0.999979138466216}], '被害人': [{'text': '徐某', 'start': 154, 'end': 156, 'probability': 0.9999752045845156}, {'text': '徐某', 'start': 121, 'end': 123, 'probability': 0.9998943832740679}, {'text': '徐某', 'start': 91, 'end': 93, 'probability': 0.9999048732152005}], '犯罪嫌疑人交通工具': [{'text': '轻型栏板货车', 'start': 38, 'end': 44, 'probability': 0.9999772311550146}], '被害人交通工具情况': [{'text': '无牌', 'start': 96, 'end': 98, 'probability': 0.9997385297185488}], '事故发生地': [{'text': '某公司路段', 'start': 70, 'end': 75, 'probability': 0.9999386079782937}], '被害人交通工具': [{'text': '三轮电动车', 'start': 98, 'end': 103, 'probability': 0.9999778271488822}]}]
83 [{'犯罪嫌疑人情况': [{'text': '未取得驾驶资格', 'start': 27, 'end': 34, 'probability': 0.9999836683896604}], '被害人': [{'text': '张某', 'start': 123, 'end': 125, 'probability': 0.9999777079771093}, {'text': '张某', 'start': 104, 'end': 106, 'probability': 0.9999504095462584}, {'text': '张某', 'start': 83, 'end': 85, 'probability': 0.9999560122536764}, {'text': '张某', 'start': 175, 'end': 177, 'probability': 0.999922634658958}], '犯罪嫌疑人交通工具': [{'text': '二轮摩托车', 'start': 41, 'end': 46, 'probability': 0.9999841452194858}], '犯罪嫌疑人交通工具情况': [{'text': '无牌证', 'start': 38, 'end': 41, 'probability': 0.9999517207213842}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 170, 'end': 174, 'probability': 0.9999682904960991}], '被害人责任认定': [{'text': '不承担此事故责任', 'start': 177, 'end': 185, 'probability': 0.9999804497432905}], '事故发生地': [{'text': '陆丰市东海镇乌坎村自来水厂路中段乡村道路', 'start': 55, 'end': 75, 'probability': 0.9999800921469273}], '被害人交通工具': [{'text': '电动自行车', 'start': 88, 'end': 93, 'probability': 0.9999853372911502}]}]
84 [{'被害人': [{'text': '金某', 'start': 63, 'end': 65, 'probability': 0.9999432571824372}, {'text': '金某', 'start': 153, 'end': 155, 'probability': 0.9999698402769752}, {'text': '金某', 'start': 80, 'end': 82, 'probability': 0.9999784232295781}], '犯罪嫌疑人交通工具': [{'text': '小客车', 'start': 35, 'end': 38, 'probability': 0.9999743701626187}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 190, 'end': 194, 'probability': 0.9999775887296636}], '事故发生地': [{'text': '自罗田县凤山镇往三里畈镇方向', 'start': 38, 'end': 52, 'probability': 0.9997124202460341}], '被害人交通工具': [{'text': '自行车', 'start': 68, 'end': 71, 'probability': 0.9999800921369797}]}]
85 [{'被害人': [{'text': '张某某', 'start': 153, 'end': 156, 'probability': 0.9370437306328796}, {'text': '张某', 'start': 113, 'end': 115, 'probability': 0.999923588288226}, {'text': '张某某', 'start': 93, 'end': 96, 'probability': 0.9999467141514629}], '犯罪嫌疑人交通工具': [{'text': '小型普通客车', 'start': 44, 'end': 50, 'probability': 0.9999874830611759}], '被害人交通工具情况': [{'text': '无牌号', 'start': 99, 'end': 102, 'probability': 0.9999496942970154}], '犯罪嫌疑人责任认定': [{'text': '主要责任', 'start': 148, 'end': 152, 'probability': 0.9999783040173753}], '被害人责任认定': [{'text': '次要责任', 'start': 159, 'end': 163, 'probability': 0.9857932837260144}], '事故发生地': [{'text': '307省道西侧机动车道', 'start': 51, 'end': 62, 'probability': 0.5316734073059948}, {'text': '十九里高速路口北侧约2公里处', 'start': 69, 'end': 83, 'probability': 0.9999413498117065}], '被害人交通工具': [{'text': '电动三轮车', 'start': 102, 'end': 107, 'probability': 0.9999827146942835}]}]
86 [{'被害人': [{'text': '王某', 'start': 146, 'end': 148, 'probability': 0.9999531512757471}, {'text': '王某', 'start': 84, 'end': 86, 'probability': 0.9999843836440334}, {'text': '王某', 'start': 78, 'end': 80, 'probability': 0.9999871254370163}], '被害人类型': [{'text': '行人', 'start': 76, 'end': 78, 'probability': 0.9999543431175084}], '犯罪嫌疑人交通工具': [{'text': '重型半挂牵引车', 'start': 34, 'end': 41, 'probability': 0.9999834299708255}], '犯罪嫌疑人责任认定': [{'text': '主要责任', 'start': 141, 'end': 145, 'probability': 0.9999772311509503}], '被害人责任认定': [{'text': '次要责任', 'start': 153, 'end': 157, 'probability': 0.999978423175861}], '事故发生地': [{'text': '法库县某某乡某某村东', 'start': 48, 'end': 58, 'probability': 0.9998940242621757}]}]
87 [{'犯罪嫌疑人情况': [{'text': '无证', 'start': 30, 'end': 32, 'probability': 0.9999761582566862}], '被害人': [{'text': '高某', 'start': 88, 'end': 90, 'probability': 0.9999803305585431}, {'text': '付某2', 'start': 198, 'end': 201, 'probability': 0.99991584000486}, {'text': '高某', 'start': 137, 'end': 139, 'probability': 0.999948263755428}, {'text': '付某2', 'start': 104, 'end': 107, 'probability': 0.9999781848010798}, {'text': '付某2', 'start': 122, 'end': 125, 'probability': 0.9999620917789116}, {'text': '付某2', 'start': 91, 'end': 94, 'probability': 0.9999864101847606}, {'text': '高某', 'start': 202, 'end': 204, 'probability': 0.9997440733781104}], '犯罪嫌疑人交通工具': [{'text': '轿车', 'start': 37, 'end': 39, 'probability': 0.9999567274016385}], '被害人责任认定': [{'text': '不承担事故责任', 'start': 210, 'end': 217, 'probability': 0.9998602908795533}], '事故发生地': [{'text': 'XX银行附近', 'start': 62, 'end': 68, 'probability': 0.9997837620474712}, {'text': '六盘水市钟山区XX路', 'start': 40, 'end': 50, 'probability': 0.6622953180313829}]}]
88 [{'犯罪嫌疑人情况': [{'text': '无证', 'start': 29, 'end': 31, 'probability': 0.9999554159898167}], '被害人': [{'text': '孙某1', 'start': 93, 'end': 96, 'probability': 0.9999862909732826}, {'text': '孙某1', 'start': 211, 'end': 214, 'probability': 0.9999390847312384}, {'text': '孙某1', 'start': 82, 'end': 85, 'probability': 0.9999899864411077}, {'text': '孙某1', 'start': 134, 'end': 137, 'probability': 0.9999880791029341}, {'text': '孙某1', 'start': 243, 'end': 246, 'probability': 0.99994516441852}, {'text': '孙某1', 'start': 152, 'end': 155, 'probability': 0.9999876022666285}, {'text': '孙某1', 'start': 142, 'end': 145, 'probability': 0.9999884367307743}], '被害人类型': [{'text': '行人', 'start': 80, 'end': 82, 'probability': 0.9999725820402006}], '犯罪嫌疑人交通工具': [{'text': '电动三轮车', 'start': 33, 'end': 38, 'probability': 0.9999762774813235}], '犯罪嫌疑人责任认定': [{'text': '主要责任', 'start': 197, 'end': 201, 'probability': 0.9999629262511576}], '被害人责任认定': [{'text': '无责任', 'start': 214, 'end': 217, 'probability': 0.9999849796796525}], '事故发生地': [{'text': '安新县徐新路山西村大富豪浴池西侧路段', 'start': 59, 'end': 77, 'probability': 0.9999767543217501}], '被害人交通工具': [{'text': '电动自行车', 'start': 116, 'end': 121, 'probability': 0.9999359856325896}]}]
89 [{'犯罪嫌疑人情况': [{'text': '酒后', 'start': 31, 'end': 33, 'probability': 0.999969721020193}], '被害人': [{'text': '李某1', 'start': 92, 'end': 95, 'probability': 0.9999859333523915}, {'text': '李某1', 'start': 73, 'end': 76, 'probability': 0.999985218102708}, {'text': '李某1', 'start': 184, 'end': 187, 'probability': 0.9999872446401241}], '犯罪嫌疑人交通工具': [{'text': '轿车', 'start': 37, 'end': 39, 'probability': 0.9902941509049512}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 249, 'end': 253, 'probability': 0.99997508536336}], '事故发生地': [{'text': '潍高路(广饶县境内)', 'start': 41, 'end': 51, 'probability': 0.9984393295538929}, {'text': '华泰总厂西门处', 'start': 58, 'end': 65, 'probability': 0.9996098346523752}], '被害人交通工具': [{'text': '电动自行车', 'start': 79, 'end': 84, 'probability': 0.9999848604585964}]}]
90 [{'被害人': [{'text': '曾某某', 'start': 89, 'end': 92, 'probability': 0.9999805689723473}, {'text': '曾某某', 'start': 60, 'end': 63, 'probability': 0.9999594692521896}, {'text': '苟某某', 'start': 171, 'end': 174, 'probability': 0.9999717475966463}, {'text': '苟某某', 'start': 101, 'end': 104, 'probability': 0.9999910593219852}, {'text': '曾某某', 'start': 167, 'end': 170, 'probability': 0.9941349365875567}, {'text': '苟某某', 'start': 118, 'end': 121, 'probability': 0.9999853373051621}], '被害人类型': [{'text': '乘坐人', 'start': 98, 'end': 101, 'probability': 0.9999774695685915}], '犯罪嫌疑人交通工具': [{'text': '重型自卸货车', 'start': 29, 'end': 35, 'probability': 0.9999710323518229}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 162, 'end': 166, 'probability': 0.999830968295214}], '被害人责任认定': [{'text': '无责任', 'start': 174, 'end': 177, 'probability': 0.999985575718739}], '事故发生地': [{'text': '秦巴大道沙溪桥附近路段', 'start': 38, 'end': 49, 'probability': 0.9999760390691392}], '被害人交通工具': [{'text': '普通二轮摩托车', 'start': 66, 'end': 73, 'probability': 0.999973177983378}]}]
91 [{'被害人': [{'text': '毛某', 'start': 89, 'end': 91, 'probability': 0.9999710323313593}, {'text': '毛某', 'start': 77, 'end': 79, 'probability': 0.9999822378921408}, {'text': '毛某', 'start': 146, 'end': 148, 'probability': 0.9999624494063255}], '犯罪嫌疑人交通工具': [{'text': '小车', 'start': 30, 'end': 32, 'probability': 0.9998457461990284}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 124, 'end': 128, 'probability': 0.999981880242558}], '事故发生地': [{'text': 'S306', 'start': 33, 'end': 37, 'probability': 0.9995048531175144}, {'text': '筻口镇集镇路段', 'start': 48, 'end': 55, 'probability': 0.999978184814637}], '被害人交通工具': [{'text': '自行车', 'start': 70, 'end': 73, 'probability': 0.9999562506025654}]}]
92 [{'被害人': [{'text': '史某', 'start': 99, 'end': 101, 'probability': 0.9999684097851684}, {'text': '史某', 'start': 110, 'end': 112, 'probability': 0.9999396809312913}, {'text': '史某', 'start': 82, 'end': 84, 'probability': 0.9999768735141288}], '犯罪嫌疑人交通工具': [{'text': '重型自卸货车', 'start': 33, 'end': 39, 'probability': 0.9999808073823715}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 141, 'end': 145, 'probability': 0.9999753237755442}], '事故发生地': [{'text': '应城市某政府2北十村路段', 'start': 50, 'end': 62, 'probability': 0.9999521976361905}], '被害人交通工具': [{'text': '二轮电动车', 'start': 89, 'end': 94, 'probability': 0.9999865293824257}]}]
93 [{'犯罪嫌疑人情况': [{'text': '酒后', 'start': 31, 'end': 33, 'probability': 0.9844608445189493}], '被害人': [{'text': '金某', 'start': 298, 'end': 300, 'probability': 0.9999109525215601}, {'text': '金某', 'start': 237, 'end': 239, 'probability': 0.9999597076332662}, {'text': '金某', 'start': 271, 'end': 273, 'probability': 0.9999302636960863}, {'text': '金某', 'start': 86, 'end': 88, 'probability': 0.9999231112253426}, {'text': '金某', 'start': 326, 'end': 328, 'probability': 0.9998067680239444}, {'text': '金某', 'start': 342, 'end': 344, 'probability': 0.9999440916122921}, {'text': '金某', 'start': 374, 'end': 376, 'probability': 0.9999468333498953}, {'text': '金某', 'start': 116, 'end': 118, 'probability': 0.9999681713690904}, {'text': '金某', 'start': 203, 'end': 205, 'probability': 0.9999554162225195}], '犯罪嫌疑人交通工具': [{'text': '小型轿车', 'start': 43, 'end': 47, 'probability': 0.9999666216728542}], '犯罪嫌疑人责任认定': [{'text': '主要责任', 'start': 198, 'end': 202, 'probability': 0.9998966476844089}], '被害人责任认定': [{'text': '次要责任', 'start': 210, 'end': 214, 'probability': 0.9999518399336438}], '事故发生地': [{'text': '阳光城市小区二期门前路段', 'start': 64, 'end': 76, 'probability': 0.9999766351133275}], '被害人交通工具': [{'text': '电动自行车', 'start': 104, 'end': 109, 'probability': 0.9999774695289716}]}]
94 [{'犯罪嫌疑人情况': [{'text': '无证', 'start': 31, 'end': 33, 'probability': 0.9999514821802507}], '被害人': [{'text': '蹇某某丽', 'start': 321, 'end': 325, 'probability': 0.9999698402379806}, {'text': '梁某某', 'start': 171, 'end': 174, 'probability': 0.9999848604749388}, {'text': '梁某某', 'start': 141, 'end': 144, 'probability': 0.9999878406880072}, {'text': '梁某某', 'start': 185, 'end': 188, 'probability': 0.9999843836347964}, {'text': '蹇某某丽', 'start': 175, 'end': 179, 'probability': 0.9999850988930348}, {'text': '蹇某某丽', 'start': 106, 'end': 110, 'probability': 0.9999879598961172}, {'text': '梁某某', 'start': 81, 'end': 84, 'probability': 0.9999852181022106}], '犯罪嫌疑人交通工具': [{'text': '小轿车', 'start': 43, 'end': 46, 'probability': 0.9999790192704836}], '被害人交通工具情况': [{'text': '无号牌', 'start': 146, 'end': 149, 'probability': 0.999915124776507}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 255, 'end': 259, 'probability': 0.9999786616310189}], '事故发生地': [{'text': '罗定市罗城街道南方电网罗定分公司门前路段', 'start': 55, 'end': 75, 'probability': 0.9999818802700702}], '被害人交通工具': [{'text': '二轮自行车', 'start': 149, 'end': 154, 'probability': 0.9999698402450861}, {'text': '普通二轮摩托车', 'start': 94, 'end': 101, 'probability': 0.9999800921220583}]}]
95 [{'犯罪嫌疑人情况': [{'text': '无证', 'start': 27, 'end': 29, 'probability': 0.99996161479244}], '被害人': [{'text': '黄某4', 'start': 276, 'end': 279, 'probability': 0.9984523594717132}, {'text': '黄某4', 'start': 239, 'end': 242, 'probability': 0.9999555353944629}, {'text': '黄某4', 'start': 124, 'end': 127, 'probability': 0.9999854565191981}], '被害人类型': [{'text': '环卫工人', 'start': 120, 'end': 124, 'probability': 0.9924000095059249}], '犯罪嫌疑人交通工具': [{'text': '二轮电动车', 'start': 39, 'end': 44, 'probability': 0.9999860525574036}], '犯罪嫌疑人交通工具情况': [{'text': '无号牌', 'start': 31, 'end': 34, 'probability': 0.9999607803875961}], '犯罪嫌疑人责任认定': [{'text': '主要责任', 'start': 271, 'end': 275, 'probability': 0.9999511246708153}], '被害人责任认定': [{'text': '次要责任', 'start': 280, 'end': 284, 'probability': 0.9999656679496525}], '事故发生地': [{'text': '中心大道万安三峰路段', 'start': 82, 'end': 92, 'probability': 0.9999778271868678}]}]
96 [{'被害人': [{'text': '周某丙', 'start': 248, 'end': 251, 'probability': 0.9998416951338527}, {'text': '周某丙', 'start': 137, 'end': 140, 'probability': 0.9999899864438362}, {'text': '周某丙', 'start': 201, 'end': 204, 'probability': 0.9999808073879421}, {'text': '周某丙', 'start': 119, 'end': 122, 'probability': 0.9999837876021047}], '犯罪嫌疑人交通工具': [{'text': '电动三轮车', 'start': 42, 'end': 47, 'probability': 0.999976277473138}], '犯罪嫌疑人交通工具情况': [{'text': '不符合安全技术标准', 'start': 32, 'end': 41, 'probability': 0.6755281445151695}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 286, 'end': 290, 'probability': 0.9999827147008205}], '事故发生地': [{'text': '淮安区翔某某大道南半幅非机动车道', 'start': 49, 'end': 65, 'probability': 0.797263295210513}, {'text': '翔某某大道与海天路路口处', 'start': 77, 'end': 89, 'probability': 0.9999593500415074}], '被害人交通工具': [{'text': '电动自行车', 'start': 130, 'end': 135, 'probability': 0.9999849796685112}]}]
97 [{'被害人': [{'text': '王某某', 'start': 48, 'end': 51, 'probability': 0.9999889135667956}, {'text': '王某某', 'start': 116, 'end': 119, 'probability': 0.9999893903998043}, {'text': '万某某', 'start': 54, 'end': 57, 'probability': 0.9999752045830519}, {'text': '徐某某', 'start': 175, 'end': 178, 'probability': 0.9467346892580437}, {'text': '王某某', 'start': 305, 'end': 308, 'probability': 0.9999556546313215}, {'text': '徐某某', 'start': 260, 'end': 263, 'probability': 0.9976466155919432}, {'text': '徐某某', 'start': 293, 'end': 296, 'probability': 0.9993294409724882}, {'text': '万某某', 'start': 309, 'end': 312, 'probability': 0.9999215614101331}, {'text': '徐某某', 'start': 199, 'end': 202, 'probability': 0.9958445781433625}, {'text': '万某某', 'start': 124, 'end': 127, 'probability': 0.9999603036061444}], '被害人类型': [{'text': '驾驶员', 'start': 150, 'end': 153, 'probability': 0.9888535738009345}], '犯罪嫌疑人交通工具': [{'text': '轻型自卸货车', 'start': 36, 'end': 42, 'probability': 0.9999835491847904}], '犯罪嫌疑人责任认定': [{'text': '主要责任', 'start': 288, 'end': 292, 'probability': 0.9999532704948138}], '被害人责任认定': [{'text': '次要责任', 'start': 300, 'end': 304, 'probability': 0.9985598719782729}, {'text': '无交通事故责任', 'start': 312, 'end': 319, 'probability': 0.9973881736056356}], '事故发生地': [{'text': '海口市美兰区白某某大道灵桂桥调头缺口路段', 'start': 64, 'end': 84, 'probability': 0.9999741317460575}]}]
98 [{'被害人': [{'text': '周某生', 'start': 174, 'end': 177, 'probability': 0.9999846220577808}, {'text': '周某生', 'start': 85, 'end': 88, 'probability': 0.9999881983151226}, {'text': '周某生', 'start': 232, 'end': 235, 'probability': 0.99996709849853}, {'text': '周某生', 'start': 145, 'end': 148, 'probability': 0.9999831915592381}, {'text': '周某生', 'start': 111, 'end': 114, 'probability': 0.9999873638520285}], '犯罪嫌疑人交通工具': [{'text': '小车', 'start': 31, 'end': 33, 'probability': 0.9998626734849125}], '被害人交通工具情况': [{'text': '无号牌', 'start': 91, 'end': 94, 'probability': 0.9999445683799024}], '犯罪嫌疑人责任认定': [{'text': '全部责任', 'start': 227, 'end': 231, 'probability': 0.9999831915467041}], '被害人责任认定': [{'text': '不负该事故的责任', 'start': 235, 'end': 243, 'probability': 0.9999829531263345}], '事故发生地': [{'text': '工业大道凯某某服装厂门前路段', 'start': 58, 'end': 72, 'probability': 0.999969244218164}], '被害人交通工具': [{'text': '三轮电动车', 'start': 94, 'end': 99, 'probability': 0.9999864101707345}]}]
99 [{'犯罪嫌疑人情况': [{'text': '酒后', 'start': 34, 'end': 36, 'probability': 0.9999753238267033}, {'text': '无证', 'start': 31, 'end': 33, 'probability': 0.9999563696092082}], '被害人': [{'text': '王某某', 'start': 47, 'end': 50, 'probability': 0.9998027703981194}, {'text': '王某某', 'start': 116, 'end': 119, 'probability': 0.9999850988942285}, {'text': '王某某', 'start': 182, 'end': 185, 'probability': 0.9999738933159961}], '被害人类型': [{'text': '车上乘员', 'start': 112, 'end': 116, 'probability': 0.9956495728043464}], '犯罪嫌疑人交通工具': [{'text': '二轮摩托车', 'start': 41, 'end': 46, 'probability': 0.9999873638528243}], '犯罪嫌疑人交通工具情况': [{'text': '无号牌', 'start': 38, 'end': 41, 'probability': 0.9999688865447922}], '犯罪嫌疑人责任认定': [{'text': '主要责任', 'start': 157, 'end': 161, 'probability': 0.9999852181010027}], '事故发生地': [{'text': '金谷嘉园小区门前东侧路段', 'start': 61, 'end': 73, 'probability': 0.9999772311498987}]}]
task_name: 刑档
[2022-10-20 22:58:24,919] [ INFO] - We are using <class 'paddlenlp.transformers.ernie.tokenizer.ErnieTokenizer'> to load 'checkpoint/model_best'.
100 [{'刑档[一档,二档,三档]': [{'text': '二档', 'probability': 0.9808830909779402}]}]
101 [{'刑档[一档,二档,三档]': [{'text': '一档', 'probability': 0.9940424114361956}]}]
102 [{'刑档[一档,二档,三档]': [{'text': '二档', 'probability': 0.9954550144631646}]}]
103 [{}]
104 [{'刑档[一档,二档,三档]': [{'text': '二档', 'probability': 0.846276048917332}]}]
105 [{'刑档[一档,二档,三档]': [{'text': '二档', 'probability': 0.9341389317186657}]}]
106 [{'刑档[一档,二档,三档]': [{'text': '一档', 'probability': 0.9981700216004583}]}]
107 [{'刑档[一档,二档,三档]': [{'text': '一档', 'probability': 0.8316393164268021}]}]
108 [{'刑档[一档,二档,三档]': [{'text': '二档', 'probability': 0.9837704851065734}]}]
109 [{'刑档[一档,二档,三档]': [{'text': '二档', 'probability': 0.9896176412692057}]}]
110 [{'刑档[一档,二档,三档]': [{'text': '一档', 'probability': 0.999462608668594}]}]
111 [{'刑档[一档,二档,三档]': [{'text': '一档', 'probability': 0.9997519401998431}]}]
112 [{'刑档[一档,二档,三档]': [{'text': '一档', 'probability': 0.9556124747794001}]}]
113 [{'刑档[一档,二档,三档]': [{'text': '一档', 'probability': 0.9994523626659984}]}]
114 [{'刑档[一档,二档,三档]': [{'text': '二档', 'probability': 0.9797007481376419}]}]
115 [{'刑档[一档,二档,三档]': [{'text': '一档', 'probability': 0.9941719020671869}]}]
116 [{'刑档[一档,二档,三档]': [{'text': '一档', 'probability': 0.993622232791946}]}]
117 [{'刑档[一档,二档,三档]': [{'text': '三档', 'probability': 0.4826780977406102}]}]
118 [{'刑档[一档,二档,三档]': [{'text': '一档', 'probability': 0.9913475188142371}]}]
119 [{'刑档[一档,二档,三档]': [{'text': '二档', 'probability': 0.9693926888549953}]}]
120 [{'刑档[一档,二档,三档]': [{'text': '一档', 'probability': 0.9989618389512529}]}]
121 [{'刑档[一档,二档,三档]': [{'text': '一档', 'probability': 0.9997712501638034}]}]
122 [{'刑档[一档,二档,三档]': [{'text': '一档', 'probability': 0.9994908020673243}]}]
123 [{'刑档[一档,二档,三档]': [{'text': '一档', 'probability': 0.9989535893651436}]}]
124 [{'刑档[一档,二档,三档]': [{'text': '三档', 'probability': 0.9821712269859404}]}]
125 [{'刑档[一档,二档,三档]': [{'text': '二档', 'probability': 0.9266373242080874}]}]
126 [{'刑档[一档,二档,三档]': [{'text': '二档', 'probability': 0.9966267219240876}]}]
127 [{'刑档[一档,二档,三档]': [{'text': '二档', 'probability': 0.9990669869410382}]}]
128 [{'刑档[一档,二档,三档]': [{'text': '二档', 'probability': 0.9992440555702586}]}]
129 [{'刑档[一档,二档,三档]': [{'text': '二档', 'probability': 0.9989724383018874}]}]
130 [{'刑档[一档,二档,三档]': [{'text': '二档', 'probability': 0.9991454989570983}]}]
131 [{'刑档[一档,二档,三档]': [{'text': '二档', 'probability': 0.990617468470326}]}]
132 [{'刑档[一档,二档,三档]': [{'text': '二档', 'probability': 0.9993149010788756}]}]
133 [{'刑档[一档,二档,三档]': [{'text': '二档', 'probability': 0.999246797664}]}]
134 [{'刑档[一档,二档,三档]': [{'text': '二档', 'probability': 0.9997176709943076}]}]
135 [{'刑档[一档,二档,三档]': [{'text': '二档', 'probability': 0.9982490459662756}]}]
136 [{'刑档[一档,二档,三档]': [{'text': '二档', 'probability': 0.9006651722641408}]}]
137 [{'刑档[一档,二档,三档]': [{'text': '二档', 'probability': 0.9992908253605997}]}]
138 [{'刑档[一档,二档,三档]': [{'text': '二档', 'probability': 0.999369658349373}]}]
139 [{'刑档[一档,二档,三档]': [{'text': '二档', 'probability': 0.9993391511177023}]}]
140 [{'刑档[一档,二档,三档]': [{'text': '二档', 'probability': 0.9992085452319373}]}]
141 [{'刑档[一档,二档,三档]': [{'text': '二档', 'probability': 0.9986949284581321}]}]
142 [{'刑档[一档,二档,三档]': [{'text': '二档', 'probability': 0.9988567884851598}]}]
143 [{'刑档[一档,二档,三档]': [{'text': '二档', 'probability': 0.996779409694927}]}]
144 [{'刑档[一档,二档,三档]': [{'text': '二档', 'probability': 0.9644375271180579}]}]
145 [{'刑档[一档,二档,三档]': [{'text': '一档', 'probability': 0.9995509801175899}]}]
146 [{'刑档[一档,二档,三档]': [{'text': '二档', 'probability': 0.997383233567632}]}]
147 [{'刑档[一档,二档,三档]': [{'text': '二档', 'probability': 0.9974167387721415}]}]
148 [{'刑档[一档,二档,三档]': [{'text': '二档', 'probability': 0.9912294772913981}]}]
149 [{'刑档[一档,二档,三档]': [{'text': '二档', 'probability': 0.9996139182323311}]}]
150 [{'刑档[一档,二档,三档]': [{'text': '二档', 'probability': 0.9963430793312185}]}]
151 [{'刑档[一档,二档,三档]': [{'text': '二档', 'probability': 0.8704550340046762}]}]
152 [{'刑档[一档,二档,三档]': [{'text': '三档', 'probability': 0.9996125405928353}]}]
153 [{'刑档[一档,二档,三档]': [{'text': '二档', 'probability': 0.9979436145197234}]}]
154 [{'刑档[一档,二档,三档]': [{'text': '三档', 'probability': 0.9987126738785719}]}]
155 [{'刑档[一档,二档,三档]': [{'text': '三档', 'probability': 0.9794621180127194}]}]
156 [{'刑档[一档,二档,三档]': [{'text': '三档', 'probability': 0.9995044986690687}]}]
157 [{'刑档[一档,二档,三档]': [{'text': '三档', 'probability': 0.9993579226550935}]}]
158 [{'刑档[一档,二档,三档]': [{'text': '二档', 'probability': 0.9983823578381017}]}]
159 [{'刑档[一档,二档,三档]': [{'text': '三档', 'probability': 0.9995021719825132}]}]
五、提交
下载result.txt并提交，目前第五名！
注：本文参考官方基线，在此基础上略有改动。
此文章为搬运
原项目链接
学大模型，用大模型上飞桨星河社区！每天8点V100G算力免费领！免费领取ERNIE 4.0 100w Token >>>
更多推荐
0
0- 0
已为社区贡献1438条内容
所有评论(0)