修正

app.py

@@ -8,18 +8,18 @@
 # %%time
 
 # ファインチューニングの実行
-python ./transformers/examples/pytorch/language-modeling/run_clm.py \
-    --model_name_or_path=rinna/japanese-gpt2-medium \
-    --train_file=natsumesouseki.txt \
-    --validation_file=natsumesouseki.txt \
-    --do_train \
-    --do_eval \
-    --num_train_epochs=3 \
-    --save_steps=5000 \
-    --save_total_limit=3 \
-    --per_device_train_batch_size=1 \
-    --per_device_eval_batch_size=1 \
-    --output_dir=output/
+# python ./transformers/examples/pytorch/language-modeling/run_clm.py \
+#    --model_name_or_path=rinna/japanese-gpt2-medium \
+#    --train_file=natsumesouseki.txt \
+#   --validation_file=natsumesouseki.txt \
+#    --do_train \
+#    --do_eval \
+#    --num_train_epochs=3 \
+#    --save_steps=5000 \
+#    --save_total_limit=3 \
+#    --per_device_train_batch_size=1 \
+#    --per_device_eval_batch_size=1 \
+#    --output_dir=output/
 
 from transformers import T5Tokenizer, AutoModelForCausalLM