End of training

all_results.json

@@ -1,14 +1,14 @@
 {
-    "epoch": 0.03,
+    "epoch": 1.0,
     "eval_loss": Infinity,
-    "eval_runtime": 12.0989,
+    "eval_runtime": 10.8674,
     "eval_samples": 301,
-    "eval_samples_per_second": 24.878,
-    "eval_steps_per_second": 3.141,
+    "eval_samples_per_second": 27.697,
+    "eval_steps_per_second": 3.497,
     "eval_wer": 1.0,
-    "train_loss": 5.443352508544922,
-    "train_runtime": 10.3804,
+    "train_loss": 4.538612365722656,
+    "train_runtime": 34.6,
     "train_samples": 704,
-    "train_samples_per_second": 1.927,
-    "train_steps_per_second": 0.963
+    "train_samples_per_second": 20.347,
+    "train_steps_per_second": 2.543
 }

eval_results.json

@@ -1,9 +1,9 @@
 {
-    "epoch": 0.03,
+    "epoch": 1.0,
     "eval_loss": Infinity,
-    "eval_runtime": 12.0989,
+    "eval_runtime": 10.8674,
     "eval_samples": 301,
-    "eval_samples_per_second": 24.878,
-    "eval_steps_per_second": 3.141,
+    "eval_samples_per_second": 27.697,
+    "eval_steps_per_second": 3.497,
     "eval_wer": 1.0
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8809cd93eb44ce44cb191d3abcf3d456d8def570fe670a5a0f3f8f57a2478541
+oid sha256:612dd692031680aae02deafa36c7fa8dadce27d4b31ce9acb75461c1e48b9ddb
 size 1247823

run.sh

@@ -4,14 +4,15 @@ python run_speech_recognition_ctc.py \
 	--dataset_config_name="ab" \
 	--output_dir="./" \
 	--overwrite_output_dir \
-	--max_steps="10" \
-	--per_device_train_batch_size="2" \
+	--num_train_epochs="1" \
+	--save_steps="100" \
+	--eval_steps="500" \
+	--per_device_train_batch_size="8" \
 	--learning_rate="3e-4" \
 	--save_total_limit="1" \
 	--evaluation_strategy="steps" \
 	--text_column_name="sentence" \
 	--length_column_name="input_length" \
-	--save_steps="5" \
 	--layerdrop="0.0" \
 	--freeze_feature_encoder \
 	--gradient_checkpointing \

special_tokens_map.json

@@ -1 +1 @@
-{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "[UNK]", "pad_token": "[PAD]", "additional_special_tokens": [{"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}]}
+{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "[UNK]", "pad_token": "[PAD]", "additional_special_tokens": [{"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}]}

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 0.03,
-    "train_loss": 5.443352508544922,
-    "train_runtime": 10.3804,
+    "epoch": 1.0,
+    "train_loss": 4.538612365722656,
+    "train_runtime": 34.6,
     "train_samples": 704,
-    "train_samples_per_second": 1.927,
-    "train_steps_per_second": 0.963
+    "train_samples_per_second": 20.347,
+    "train_steps_per_second": 2.543
 }

trainer_state.json

@@ -1,25 +1,25 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.028409090909090908,
-  "global_step": 10,
+  "epoch": 1.0,
+  "global_step": 88,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.03,
-      "step": 10,
-      "total_flos": 5622472894464.0,
-      "train_loss": 5.443352508544922,
-      "train_runtime": 10.3804,
-      "train_samples_per_second": 1.927,
-      "train_steps_per_second": 0.963
+      "epoch": 1.0,
+      "step": 88,
+      "total_flos": 130613499224064.0,
+      "train_loss": 4.538612365722656,
+      "train_runtime": 34.6,
+      "train_samples_per_second": 20.347,
+      "train_steps_per_second": 2.543
     }
   ],
-  "max_steps": 10,
+  "max_steps": 88,
   "num_train_epochs": 1,
-  "total_flos": 5622472894464.0,
+  "total_flos": 130613499224064.0,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:179b27dba52e10bae586a6a067bfb01f31c209e263abb88cb96bee05b1b0c8dd
+oid sha256:20957f7464463ef438ad81644f984c525e32f342ea8286c481883de4c457daef
 size 2991