测试报错

[YYD1319]

Reminder

• I have read the README and searched the existing issues.

Reproduction

CUDA_VISIBLE_DEVICES=1 python /home/yyd/code/chat/LLaMA-Factory_231116/src/train_bash.py
--stage sft
--model_name_or_path /home/yyd/code/chat/models/ChatGLM3-6B
--do_predict True
--dataset_dir /home/yyd/code/chat/LLaMA-Factory_231116/data
--dataset steel_QA_test_2661
--template chatglm3
--finetuning_type lora
--checkpoint_dir $checkpoint_dir
--output_dir $output_dir
--per_device_eval_batch_size 4
--max_samples 100
--cutoff_len 256
--max_new_tokens 256
--predict_with_generate True

Expected behavior

No response

System Info

No response

Others

Traceback (most recent call last):
File "/home/yyd/code/chat/LLaMA-Factory_231116/src/train_bash.py", line 14, in
main()
File "/home/yyd/code/chat/LLaMA-Factory_231116/src/train_bash.py", line 5, in main
run_exp()
File "/home/yyd/code/chat/LLaMA-Factory_231116/src/llmtuner/train/tuner.py", line 26, in run_exp
run_sft(model_args, data_args, training_args, finetuning_args, generating_args, callbacks)
File "/home/yyd/code/chat/LLaMA-Factory_231116/src/llmtuner/train/sft/workflow.py", line 85, in run_sft
predict_results = trainer.predict(dataset, metric_key_prefix="predict", **gen_kwargs)
File "/home/yyd/anaconda3/envs/llama_factory_2/lib/python3.10/site-packages/transformers/trainer_seq2seq.py", line 228, in predict
return super().predict(test_dataset, ignore_keys=ignore_keys, metric_key_prefix=metric_key_prefix)
File "/home/yyd/anaconda3/envs/llama_factory_2/lib/python3.10/site-packages/transformers/trainer.py", line 3142, in predict
output = eval_loop(
File "/home/yyd/anaconda3/envs/llama_factory_2/lib/python3.10/site-packages/transformers/trainer.py", line 3359, in evaluation_loop
metrics = self.compute_metrics(EvalPrediction(predictions=all_preds, label_ids=all_labels))
File "/home/yyd/code/chat/LLaMA-Factory_231116/src/llmtuner/train/sft/metric.py", line 51, in call
rouge = Rouge()
NameError: name 'Rouge' is not defined. Did you mean: 'rouge'?

这一段代码为什么会报错呀：
for pred, label in zip(decoded_preds, decoded_labels):
hypothesis = list(jieba.cut(pred))
reference = list(jieba.cut(label))

        if len(" ".join(hypothesis).split()) == 0 or len(" ".join(reference).split()) == 0:
            result = {"rouge-1": {"f": 0.0}, "rouge-2": {"f": 0.0}, "rouge-l": {"f": 0.0}}
        else:
            rouge = Rouge()
            scores = rouge.get_scores(" ".join(hypothesis), " ".join(reference))
            result = scores[0]

[YYD1319]

是不是有bug，这里应该改一下？
./src/llmtuner/extras/packages.py
21: _rouge_available = is_package_available("rouge-chinese") ->
_rouge_available = is_package_available("rouge_chinese")

[hiyouga]

已修复