Training in progress, step 100

.ipynb_checkpoints/run-checkpoint.sh

@@ -1,20 +1,22 @@
 python run_speech_recognition_ctc.py \
 	--dataset_name="mozilla-foundation/common_voice_7_0" \
-	--model_name_or_path="facebook/wav2vec2-xls-r-300m" \
+	--model_name_or_path="facebook/wav2vec2-large-xlsr-53" \
 	--dataset_config_name="ur" \
 	--output_dir="./" \
 	--overwrite_output_dir \
-	--max_steps="2500" \
-	--per_device_train_batch_size="16" \
+	--max_steps="2000" \
+	--gradient_accumulation_steps="2" \
+	--lr_scheduler_type="linear" \
+	--per_device_train_batch_size="64" \
 	--learning_rate="3e-4" \
-	--save_total_limit="10" \
+	--save_total_limit="2" \
 	--evaluation_strategy="steps" \
 	--text_column_name="sentence" \
 	--length_column_name="input_length" \
 	--chars_to_ignore ,\?\.\!\-\;\:\"\“\%\‘\”\�\ّ\ٔ\َِ,ً\
 	--save_steps="100" \
 	--layerdrop="0.0" \
-	--num_train_epochs="500.0" \
+	--num_train_epochs="750.0" \
 	--freeze_feature_encoder \
 	--gradient_checkpointing \
 	--fp16 \

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "facebook/wav2vec2-xls-r-300m",
+  "_name_or_path": "facebook/wav2vec2-large-xlsr-53",
   "activation_dropout": 0.0,
   "adapter_kernel_size": 3,
   "adapter_stride": 2,
@@ -59,12 +59,20 @@
   "intermediate_size": 4096,
   "layer_norm_eps": 1e-05,
   "layerdrop": 0.0,
+  "mask_channel_length": 10,
+  "mask_channel_min_space": 1,
+  "mask_channel_other": 0.0,
+  "mask_channel_prob": 0.0,
+  "mask_channel_selection": "static",
   "mask_feature_length": 10,
   "mask_feature_min_masks": 0,
   "mask_feature_prob": 0.0,
   "mask_time_length": 10,
   "mask_time_min_masks": 2,
+  "mask_time_min_space": 1,
+  "mask_time_other": 0.0,
   "mask_time_prob": 0.05,
+  "mask_time_selection": "static",
   "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:a0cb80392e2e0813418bf0b7f1b4abd1be1a7983385a140c24a112556e1bf723
+oid sha256:db6bb3faa65845292b326eba507ccb53f867819da6892213cd1f606491f870a6
 size 1262149169

run.sh

@@ -1,20 +1,22 @@
 python run_speech_recognition_ctc.py \
 	--dataset_name="mozilla-foundation/common_voice_7_0" \
-	--model_name_or_path="facebook/wav2vec2-xls-r-300m" \
+	--model_name_or_path="facebook/wav2vec2-large-xlsr-53" \
 	--dataset_config_name="ur" \
 	--output_dir="./" \
 	--overwrite_output_dir \
-	--max_steps="2500" \
-	--per_device_train_batch_size="16" \
+	--max_steps="2000" \
+	--gradient_accumulation_steps="2" \
+	--lr_scheduler_type="linear" \
+	--per_device_train_batch_size="64" \
 	--learning_rate="3e-4" \
-	--save_total_limit="10" \
+	--save_total_limit="2" \
 	--evaluation_strategy="steps" \
 	--text_column_name="sentence" \
 	--length_column_name="input_length" \
 	--chars_to_ignore ,\?\.\!\-\;\:\"\“\%\‘\”\�\ّ\ٔ\َِ,ً\
 	--save_steps="100" \
 	--layerdrop="0.0" \
-	--num_train_epochs="500.0" \
+	--num_train_epochs="750.0" \
 	--freeze_feature_encoder \
 	--gradient_checkpointing \
 	--fp16 \

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}, {"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 sha256:60e6c9e0546095eca72327a4e5b808d6df6823b0e2b458beaf565177635dba6e
+oid sha256:19a3ea893e13d240e35254a9b7643af00427f3c37f5ec79d636fc7b249979609
 size 2991