Training in progress, step 50

.ipynb_checkpoints/run-checkpoint.sh

@@ -4,19 +4,19 @@ python run_speech_recognition_ctc.py \
 	--dataset_config_name="sv-SE" \
 	--output_dir="./" \
 	--overwrite_output_dir \
-	--num_train_epochs="1" \
+	--num_train_epochs="2" \
 	--per_device_train_batch_size="8" \
 	--per_device_eval_batch_size="8" \
 	--gradient_accumulation_steps="4" \
 	--learning_rate="7.5e-5" \
-	--warmup_steps="2000" \
+	--warmup_steps="20" \
 	--length_column_name="input_length" \
 	--evaluation_strategy="steps" \
 	--text_column_name="sentence" \
 	--chars_to_ignore , ? . ! \- \; \: \" “ % ‘ ” � — ’ … – \
-	--save_steps="500" \
-	--eval_steps="500" \
-	--logging_steps="100" \
+	--save_steps="50" \
+	--eval_steps="50" \
+	--logging_steps="10" \
 	--layerdrop="0.0" \
 	--activation_dropout="0.1" \
 	--save_total_limit="3" \

checkpoint-50/config.json

@@ -0,0 +1,107 @@
+{
+  "_name_or_path": "facebook/wav2vec2-xls-r-300m",
+  "activation_dropout": 0.1,
+  "adapter_kernel_size": 3,
+  "adapter_stride": 2,
+  "add_adapter": false,
+  "apply_spec_augment": true,
+  "architectures": [
+    "Wav2Vec2ForCTC"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "classifier_proj_size": 256,
+  "codevector_dim": 768,
+  "contrastive_logits_temperature": 0.1,
+  "conv_bias": true,
+  "conv_dim": [
+    512,
+    512,
+    512,
+    512,
+    512,
+    512,
+    512
+  ],
+  "conv_kernel": [
+    10,
+    3,
+    3,
+    3,
+    3,
+    2,
+    2
+  ],
+  "conv_stride": [
+    5,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2
+  ],
+  "ctc_loss_reduction": "mean",
+  "ctc_zero_infinity": false,
+  "diversity_loss_weight": 0.1,
+  "do_stable_layer_norm": true,
+  "eos_token_id": 2,
+  "feat_extract_activation": "gelu",
+  "feat_extract_dropout": 0.0,
+  "feat_extract_norm": "layer",
+  "feat_proj_dropout": 0.0,
+  "feat_quantizer_dropout": 0.0,
+  "final_dropout": 0.0,
+  "hidden_act": "gelu",
+  "hidden_dropout": 0.0,
+  "hidden_size": 1024,
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "layer_norm_eps": 1e-05,
+  "layerdrop": 0.0,
+  "mask_feature_length": 64,
+  "mask_feature_min_masks": 0,
+  "mask_feature_prob": 0.25,
+  "mask_time_length": 10,
+  "mask_time_min_masks": 2,
+  "mask_time_prob": 0.75,
+  "model_type": "wav2vec2",
+  "num_adapter_layers": 3,
+  "num_attention_heads": 16,
+  "num_codevector_groups": 2,
+  "num_codevectors_per_group": 320,
+  "num_conv_pos_embedding_groups": 16,
+  "num_conv_pos_embeddings": 128,
+  "num_feat_extract_layers": 7,
+  "num_hidden_layers": 24,
+  "num_negatives": 100,
+  "output_hidden_size": 1024,
+  "pad_token_id": 34,
+  "proj_codevector_dim": 768,
+  "tdnn_dilation": [
+    1,
+    2,
+    3,
+    1,
+    1
+  ],
+  "tdnn_dim": [
+    512,
+    512,
+    512,
+    512,
+    1500
+  ],
+  "tdnn_kernel": [
+    5,
+    3,
+    3,
+    1,
+    1
+  ],
+  "torch_dtype": "float32",
+  "transformers_version": "4.16.0.dev0",
+  "use_weighted_layer_sum": false,
+  "vocab_size": 37,
+  "xvector_output_dim": 512
+}

checkpoint-50/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e59c74b41c4b1b3768c9b03acbfe97aeac7e3a7fee68c17a27fc0f6fc4005304
+size 2490361937

checkpoint-50/preprocessor_config.json

@@ -0,0 +1,9 @@
+{
+  "do_normalize": true,
+  "feature_extractor_type": "Wav2Vec2FeatureExtractor",
+  "feature_size": 1,
+  "padding_side": "right",
+  "padding_value": 0,
+  "return_attention_mask": true,
+  "sampling_rate": 16000
+}

checkpoint-50/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3dcafa9a58adee9c2dd0c4c04b15f3b1affea9b4fad06e5c8af956a938054769
+size 1262075377

checkpoint-50/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:14dd13f385e830f5e013aa12c8f455e60dbd26d2b24429728757725e99609101
+size 14567

checkpoint-50/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:58db19cd158cab87b95a3329bd53321ccb26cb15fcfe671116e04cc3c9745427
+size 559

checkpoint-50/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fb707a581933f4f3461df9727f0a85dd78930b27f2451939d060d737f2c9f1d2
+size 623

checkpoint-50/trainer_state.json

@@ -0,0 +1,55 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.145032632342277,
+  "global_step": 50,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.03,
+      "learning_rate": 3.75e-05,
+      "loss": 12.1562,
+      "step": 10
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 7.125e-05,
+      "loss": 8.7679,
+      "step": 20
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 7.398952095808383e-05,
+      "loss": 5.3683,
+      "step": 30
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 7.286676646706586e-05,
+      "loss": 4.3219,
+      "step": 40
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 7.17440119760479e-05,
+      "loss": 3.7182,
+      "step": 50
+    },
+    {
+      "epoch": 0.15,
+      "eval_loss": 3.836604595184326,
+      "eval_runtime": 133.4846,
+      "eval_samples_per_second": 34.611,
+      "eval_steps_per_second": 4.33,
+      "eval_wer": 1.0,
+      "step": 50
+    }
+  ],
+  "max_steps": 688,
+  "num_train_epochs": 2,
+  "total_flos": 1.4741265581374464e+17,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-50/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:77806a16ec6ef209c8a5c1c085159a0bc8e45c23f2d143f2c13e01527f13b5b2
+size 2991

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7fc2f814275f65fd80d24f657344d3cf4a8a27f2158ee4f63e194a2e441bc0bf
+oid sha256:3dcafa9a58adee9c2dd0c4c04b15f3b1affea9b4fad06e5c8af956a938054769
 size 1262075377

run.sh

@@ -4,19 +4,19 @@ python run_speech_recognition_ctc.py \
 	--dataset_config_name="sv-SE" \
 	--output_dir="./" \
 	--overwrite_output_dir \
-	--num_train_epochs="1" \
+	--num_train_epochs="2" \
 	--per_device_train_batch_size="8" \
 	--per_device_eval_batch_size="8" \
 	--gradient_accumulation_steps="4" \
 	--learning_rate="7.5e-5" \
-	--warmup_steps="2000" \
+	--warmup_steps="20" \
 	--length_column_name="input_length" \
 	--evaluation_strategy="steps" \
 	--text_column_name="sentence" \
 	--chars_to_ignore , ? . ! \- \; \: \" “ % ‘ ” � — ’ … – \
-	--save_steps="500" \
-	--eval_steps="500" \
-	--logging_steps="100" \
+	--save_steps="50" \
+	--eval_steps="50" \
+	--logging_steps="10" \
 	--layerdrop="0.0" \
 	--activation_dropout="0.1" \
 	--save_total_limit="3" \

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}]}
+{"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:c42eb9a8cea359c1adc4bcd28833b79932fb4a6db79a23dfc44abef42675022a
+oid sha256:77806a16ec6ef209c8a5c1c085159a0bc8e45c23f2d143f2c13e01527f13b5b2
 size 2991