Training in progress, step 500

.ipynb_checkpoints/run-checkpoint.sh

@@ -4,11 +4,11 @@ python run_speech_recognition_ctc.py \
         --dataset_config_name="pa-IN" \
         --output_dir="./" \
         --overwrite_output_dir \
-        --num_train_epochs="200" \
+        --num_train_epochs="100" \
         --per_device_train_batch_size="16" \
         --per_device_eval_batch_size="8" \
         --learning_rate="3e-4" \
-        --warmup_steps="1500" \
+        --warmup_steps="1200" \
         --length_column_name="input_length" \
         --evaluation_strategy="steps" \
         --text_column_name="sentence" \
@@ -20,9 +20,9 @@ python run_speech_recognition_ctc.py \
         --save_total_limit="1" \
         --freeze_feature_encoder \
         --feat_proj_dropout="0.0" \
-        --mask_time_prob="0.75" \
+        --mask_time_prob="0.05" \
         --mask_time_length="10" \
-        --mask_feature_prob="0.25" \
+        --mask_feature_prob="0.05" \
         --mask_feature_length="64" \
         --chars_to_ignore , ? . ! \- \; \: \" “ % ‘ ” � — ’ … – « » „ \` _ \
         --gradient_checkpointing \

config.json

@@ -61,10 +61,10 @@
   "layerdrop": 0.0,
   "mask_feature_length": 64,
   "mask_feature_min_masks": 0,
-  "mask_feature_prob": 0.25,
+  "mask_feature_prob": 0.05,
   "mask_time_length": 10,
   "mask_time_min_masks": 2,
-  "mask_time_prob": 0.75,
+  "mask_time_prob": 0.05,
   "model_type": "wav2vec2",
   "num_adapter_layers": 3,
   "num_attention_heads": 16,

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8c4b7517f11637e6acabdd9728be577c8a021ad5019ebc00b7863a59167e474b
+oid sha256:f3047206709c8430e48b6d738fa94dc80544d7a168c36e603faa2e4e22f5106b
 size 1262194289

run.sh

@@ -4,11 +4,11 @@ python run_speech_recognition_ctc.py \
         --dataset_config_name="pa-IN" \
         --output_dir="./" \
         --overwrite_output_dir \
-        --num_train_epochs="200" \
+        --num_train_epochs="100" \
         --per_device_train_batch_size="16" \
         --per_device_eval_batch_size="8" \
         --learning_rate="3e-4" \
-        --warmup_steps="1500" \
+        --warmup_steps="1200" \
         --length_column_name="input_length" \
         --evaluation_strategy="steps" \
         --text_column_name="sentence" \
@@ -20,9 +20,9 @@ python run_speech_recognition_ctc.py \
         --save_total_limit="1" \
         --freeze_feature_encoder \
         --feat_proj_dropout="0.0" \
-        --mask_time_prob="0.75" \
+        --mask_time_prob="0.05" \
         --mask_time_length="10" \
-        --mask_feature_prob="0.25" \
+        --mask_feature_prob="0.05" \
         --mask_feature_length="64" \
         --chars_to_ignore , ? . ! \- \; \: \" “ % ‘ ” � — ’ … – « » „ \` _ \
         --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}]}
+{"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

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7ce5521e7d5bfb2e376e94f2b63e11df51adac0d076e157e0d59b86829b64f4e
+oid sha256:6b1aedc4fe7ec92200eafc3ebd49f8423ed1f76a39299859626edb51c651c948
 size 2991