LEVEN

LEVEN copied to clipboard

我在执行Baselines中BERT的时候，运行run_train.sh报错： `$ CUDA_VISIBLE_DEVICES=0,1,2,3 python run_leven.py --data_dir ./data --model_type bert --output_dir ./saved --max_seq_length 512 --per_gpu_train_batch_size 8--per_gpu_eval_batch_size 4--gradient_accumulation_steps 2--learning_rate 5e-5--num_train_epochs 4--save_steps 100--logging_steps 100--seed 100--do_train --do_eval --eval_all_checkpoints --overwrite_output_dir Traceback (most recent call last):...

ROY-SNN

# 出现问题的模块 ./Baselines/BERT # 环境 CUDA：10.1 其余版本和类库按照README.md的Requirements配置。 # 表现 在运行到run_leven.py中加载NER模型的时候，会出现Segmentation fault (core dumped)。 ```python ner_model = model_class.from_pretrained( args.model_name_or_path, from_tf=bool(".ckpt" in args.model_name_or_path), config=config, cache_dir=args.cache_dir if args.cache_dir else None, ) ``` #...

IssacStudent

Zfade

hello,i have some confusions in the performance in baselines in LED on LEVEN. the performance listed is the averages and standard deviations .  as we can see ,the deviations...

Yyy11181

### 请问中国判决文书的网站访问不是很流畅，怎么大规模获取到原始的判决文书数据集的呢？

6

guocxian