错误运行 run_seq2seq.py 变形金刚训练脚本

Question

我正在尝试训练 seq2seq 模型。我运行 Colab 中的示例代码：

!git clone https://github.com/huggingface/transformers
!git clone https://github.com/huggingface/datasets
!pip install transformers
!pip install datasets

!python transformers/examples/seq2seq/run_seq2seq.py \
    --model_name_or_path t5-small \
    --do_train \
    --do_eval \
    --task summarization \
    --dataset_name xsum \
    --output_dir /tmp/tst-summarization \
    --per_device_train_batch_size=4 \
    --per_device_eval_batch_size=4 \
    --overwrite_output_dir \
    --predict_with_generate \
    --max_train_samples 500 \
    --max_val_samples 500

遇到这个错误

I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0
Traceback (most recent call last):
  File "transformers/examples/seq2seq/run_seq2seq.py", line 47, in <module>
    from transformers.file_utils import is_offline_mode
ImportError: cannot import name 'is_offline_mode' from 'transformers.file_utils' (/usr/local/lib/python3.7/dist-packages/transformers/file_utils.py)

有什么想法吗？

Answer 1

问题是您克隆了存储库的 master 分支，并尝试运行 run_seq2seq.py 脚本，其中包含该 master 分支后面的 transformers 版本 (4.3.3)。

run_seq2seq.py 已更新为在 3 月 6 日导入 is_offline_mode merge。

您需要做的就是克隆用于您使用的变形金刚版本的分支：

!git clone --branch v4.3.3-release https://github.com/huggingface/transformers

P.S.: 我认为你不需要克隆数据集库。

错误运行 run_seq2seq.py 变形金刚训练脚本

Error running run_seq2seq.py Transformers training script

python

nlp

machine-learning

tensorflow

huggingface-transformers

错误 运行 run_seq2seq.py 变形金刚训练脚本

Error running run_seq2seq.py Transformers training script

python

nlp

machine-learning

tensorflow

huggingface-transformers

错误运行 run_seq2seq.py 变形金刚训练脚本