Training in progress, step 500

config.json

@@ -6,7 +6,7 @@
   "add_adapter": false,
   "apply_spec_augment": true,
   "architectures": [
-    "Wav2Vec2ForPreTraining"
+    "Wav2Vec2ForCTC"
   ],
   "attention_dropout": 0.0,
   "bos_token_id": 1,
@@ -64,7 +64,7 @@
   "mask_feature_prob": 0.25,
   "mask_time_length": 10,
   "mask_time_min_masks": 2,
-  "mask_time_prob": 0.75,
+  "mask_time_prob": 0.4,
   "model_type": "wav2vec2",
   "num_adapter_layers": 3,
   "num_attention_heads": 16,

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9b0307eb3c0a9438759535fcfcc104d7b0740e05e6f3411660bf1adaa0f99b3a
+size 8637973773

run_bnb.sh

@@ -9,7 +9,7 @@ python run_speech_recognition_ctc_bnb.py \
 --per_device_train_batch_size="4" \
 --per_device_eval_batch_size="4" \
 --gradient_accumulation_steps="8" \
---learning_rate="3e-6" \
+--learning_rate="1e-6" \
 --warmup_steps="2000" \
 --length_column_name="input_length" \
 --evaluation_strategy="steps" \
@@ -23,7 +23,7 @@ python run_speech_recognition_ctc_bnb.py \
 --save_total_limit="2" \
 --freeze_feature_encoder \
 --feat_proj_dropout="0.0" \
---mask_time_prob="0.75" \
+--mask_time_prob="0.4" \
 --mask_time_length="10" \
 --mask_feature_prob="0.25" \
 --mask_feature_length="64" \
@@ -31,4 +31,5 @@ python run_speech_recognition_ctc_bnb.py \
 --use_auth_token \
 --fp16 \
 --group_by_length \
---do_train --do_eval
+--do_train --do_eval \
+--push_to_hub

runs/Jan30_00-36-51_job-e3562c4a-f6a3-47dc-8131-7e578f1139a3/1643503156.416175/events.out.tfevents.1643503156.job-e3562c4a-f6a3-47dc-8131-7e578f1139a3.1780523.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b5ab45cebf194e6c08cba95c3b23cb71c5e3270eb29ccfb1564ee0752c6d00cb
+size 4753

runs/Jan30_00-36-51_job-e3562c4a-f6a3-47dc-8131-7e578f1139a3/events.out.tfevents.1643503156.job-e3562c4a-f6a3-47dc-8131-7e578f1139a3.1780523.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:39f0fe242ca5001faca64838667cbdabb3ca112fad4ca92f511fb4d08c87580e
+size 12811

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}]}
+{"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}]}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b7ff6601b1bceeeee90202d6ca36380acbd3f91cd455add453e82e644446efb5
+size 2991