Model save

all_results.json

@@ -7,7 +7,7 @@
     "eval_steps_per_second": 2.803,
     "total_flos": 9.82152667464321e+19,
     "train_loss": 0.0,
-    "train_runtime": 0.1001,
-    "train_samples_per_second": 12657259.542,
-    "train_steps_per_second": 98926.975
+    "train_runtime": 0.0755,
+    "train_samples_per_second": 16783731.897,
+    "train_steps_per_second": 131178.777
 }

command

@@ -1 +1 @@
-python run_image_classification_val_test.py     --train_dir /trunk/shared/eebo_data/images_cropped/     --validation_dir /trunk/shared/eebo_data/images_cropped/     --output_dir ./test_microsoft_dit/     --remove_unused_columns False     --label_column_name labels     --do_train     --do_eval     --push_to_hub     --push_to_hub_model_id donut-base-beans     --learning_rate 3e-5     --num_train_epochs 5     --per_device_train_batch_size 32     --per_device_eval_batch_size 32     --logging_strategy steps     --logging_steps 10     --eval_strategy steps     --save_strategy steps     --load_best_model_at_end True     --save_total_limit 5     --seed 1337     --ignore_mismatched_sizes     --image_column_name image_url     --label_column_name label     --model_name_or_path microsoft/dit-base-finetuned-rvlcdip     --resume_from_checkpoint /trunk2/yufei/summer24/transformers/examples/pytorch/image-classification/test_microsoft_dit/checkpoint-9905
+python run_image_classification_val_test.py     --train_dir /trunk/shared/eebo_data/images_cropped/     --validation_dir /trunk/shared/eebo_data/images_cropped/     --output_dir ./test_microsoft_dit/     --remove_unused_columns False     --label_column_name labels     --do_train     --do_eval     --push_to_hub     --push_to_hub_model_id donut-base-beans     --learning_rate 3e-5     --num_train_epochs 5     --per_device_train_batch_size 32     --per_device_eval_batch_size 32     --logging_strategy steps     --logging_steps 10     --eval_strategy epoch     --save_strategy epoch     --load_best_model_at_end True     --save_total_limit 5     --seed 1337     --ignore_mismatched_sizes     --image_column_name image_url     --label_column_name label     --model_name_or_path microsoft/dit-base-finetuned-rvlcdip     --resume_from_checkpoint /trunk2/yufei/summer24/transformers/examples/pytorch/image-classification/test_microsoft_dit/checkpoint-9905

train_results.json

@@ -2,7 +2,7 @@
     "epoch": 5.0,
     "total_flos": 9.82152667464321e+19,
     "train_loss": 0.0,
-    "train_runtime": 0.1001,
-    "train_samples_per_second": 12657259.542,
-    "train_steps_per_second": 98926.975
+    "train_runtime": 0.0755,
+    "train_samples_per_second": 16783731.897,
+    "train_steps_per_second": 131178.777
 }

trainer_state.json

@@ -6988,9 +6988,9 @@
       "step": 9905,
       "total_flos": 9.82152667464321e+19,
       "train_loss": 0.0,
-      "train_runtime": 0.1001,
-      "train_samples_per_second": 12657259.542,
-      "train_steps_per_second": 98926.975
+      "train_runtime": 0.0755,
+      "train_samples_per_second": 16783731.897,
+      "train_steps_per_second": 131178.777
     }
   ],
   "logging_steps": 10,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a5b65dfdcf6333961fd335451fa9ff0ffadf6e07ca172ab650ee52f3df8a8fee
-size 5240
+oid sha256:ba1219e59ef01629bdf0cd0e29d373c314b384e2f22eb4b457e08427db6eec5c
+size 5368