Training in progress, epoch 8, checkpoint

checkpoint-2064/config.json

@@ -0,0 +1,82 @@
+{
+  "_name_or_path": "facebook/w2v-bert-2.0",
+  "activation_dropout": 0.0,
+  "adapter_act": "relu",
+  "adapter_kernel_size": 3,
+  "adapter_stride": 2,
+  "add_adapter": true,
+  "apply_spec_augment": false,
+  "architectures": [
+    "Wav2Vec2BertForCTC"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "classifier_proj_size": 768,
+  "codevector_dim": 768,
+  "conformer_conv_dropout": 0.1,
+  "contrastive_logits_temperature": 0.1,
+  "conv_depthwise_kernel_size": 31,
+  "ctc_loss_reduction": "mean",
+  "ctc_zero_infinity": true,
+  "diversity_loss_weight": 0.1,
+  "eos_token_id": 2,
+  "feat_proj_dropout": 0.0,
+  "feat_quantizer_dropout": 0.0,
+  "feature_projection_input_dim": 160,
+  "final_dropout": 0.1,
+  "hidden_act": "swish",
+  "hidden_dropout": 0.0,
+  "hidden_size": 1024,
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "layer_norm_eps": 1e-05,
+  "layerdrop": 0.0,
+  "left_max_position_embeddings": 64,
+  "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_prob": 0.0,
+  "max_source_positions": 5000,
+  "model_type": "wav2vec2-bert",
+  "num_adapter_layers": 1,
+  "num_attention_heads": 16,
+  "num_codevector_groups": 2,
+  "num_codevectors_per_group": 320,
+  "num_hidden_layers": 24,
+  "num_negatives": 100,
+  "output_hidden_size": 1024,
+  "pad_token_id": 43,
+  "position_embeddings_type": "relative_key",
+  "proj_codevector_dim": 768,
+  "right_max_position_embeddings": 8,
+  "rotary_embedding_base": 10000,
+  "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": "4.46.0",
+  "use_intermediate_ffn_before_adapter": false,
+  "use_weighted_layer_sum": false,
+  "vocab_size": 46,
+  "xvector_output_dim": 512
+}

checkpoint-2064/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4723c2892ab0b67a9a0bf859c25a50c1d8660cbbb6f0b927f0d5bc44f50e34ff
+size 2423003160

checkpoint-2064/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:95c06034a9fa258ea44317a56c70d5a0a161fbcf03cea611e529401be3cf4564
+size 4846467050

checkpoint-2064/preprocessor_config.json

@@ -0,0 +1,11 @@
+{
+  "feature_extractor_type": "SeamlessM4TFeatureExtractor",
+  "feature_size": 80,
+  "num_mel_bins": 80,
+  "padding_side": "right",
+  "padding_value": 1,
+  "processor_class": "Wav2Vec2BertProcessor",
+  "return_attention_mask": true,
+  "sampling_rate": 16000,
+  "stride": 2
+}

checkpoint-2064/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3946a5be162eddc3adb0d0cf4b6d46b1345ebdd7ba4068542c664439392a1a8d
+size 14244

checkpoint-2064/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:182ffb058950d6e24bc33f00c19379567da3395b6ddcece3c03d048ea96cb7b7
+size 1064

checkpoint-2064/trainer_state.json

@@ -0,0 +1,178 @@
+{
+  "best_metric": 0.4603155629930672,
+  "best_model_checkpoint": "w2v-bert-2.0-CV_Fleurs-lg-5hrs-v4/checkpoint-2064",
+  "epoch": 8.0,
+  "eval_steps": 500,
+  "global_step": 2064,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": 11.246596336364746,
+      "learning_rate": 5.1200000000000004e-05,
+      "loss": 1.9959,
+      "step": 258
+    },
+    {
+      "epoch": 1.0,
+      "eval_cer": 0.13379445701894616,
+      "eval_loss": 0.5475470423698425,
+      "eval_runtime": 29.6601,
+      "eval_samples_per_second": 27.849,
+      "eval_steps_per_second": 6.979,
+      "eval_wer": 0.6078173559646187,
+      "step": 258
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.1338785886764526,
+      "learning_rate": 9.994466403162056e-05,
+      "loss": 0.5447,
+      "step": 516
+    },
+    {
+      "epoch": 2.0,
+      "eval_cer": 0.1309612371199291,
+      "eval_loss": 0.5649892687797546,
+      "eval_runtime": 29.1014,
+      "eval_samples_per_second": 28.384,
+      "eval_steps_per_second": 7.113,
+      "eval_wer": 0.5978962467128854,
+      "step": 516
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 5.585188388824463,
+      "learning_rate": 9.892490118577076e-05,
+      "loss": 0.4574,
+      "step": 774
+    },
+    {
+      "epoch": 3.0,
+      "eval_cer": 0.11218917678342487,
+      "eval_loss": 0.45441317558288574,
+      "eval_runtime": 29.5398,
+      "eval_samples_per_second": 27.962,
+      "eval_steps_per_second": 7.007,
+      "eval_wer": 0.5168539325842697,
+      "step": 774
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 3.5034706592559814,
+      "learning_rate": 9.790513833992096e-05,
+      "loss": 0.3712,
+      "step": 1032
+    },
+    {
+      "epoch": 4.0,
+      "eval_cer": 0.10859621076623562,
+      "eval_loss": 0.4465671181678772,
+      "eval_runtime": 29.2068,
+      "eval_samples_per_second": 28.281,
+      "eval_steps_per_second": 7.087,
+      "eval_wer": 0.5049007889074827,
+      "step": 1032
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 17.74321174621582,
+      "learning_rate": 9.688932806324111e-05,
+      "loss": 0.3187,
+      "step": 1290
+    },
+    {
+      "epoch": 5.0,
+      "eval_cer": 0.10036562781937036,
+      "eval_loss": 0.44379866123199463,
+      "eval_runtime": 29.231,
+      "eval_samples_per_second": 28.258,
+      "eval_steps_per_second": 7.082,
+      "eval_wer": 0.4769304327038011,
+      "step": 1290
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 10.85033893585205,
+      "learning_rate": 9.586956521739131e-05,
+      "loss": 0.276,
+      "step": 1548
+    },
+    {
+      "epoch": 6.0,
+      "eval_cer": 0.1056996786907042,
+      "eval_loss": 0.4188626706600189,
+      "eval_runtime": 29.2483,
+      "eval_samples_per_second": 28.241,
+      "eval_steps_per_second": 7.077,
+      "eval_wer": 0.4794405928759264,
+      "step": 1548
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 1.633737564086914,
+      "learning_rate": 9.484980237154151e-05,
+      "loss": 0.235,
+      "step": 1806
+    },
+    {
+      "epoch": 7.0,
+      "eval_cer": 0.1004447680400133,
+      "eval_loss": 0.4695667326450348,
+      "eval_runtime": 29.6156,
+      "eval_samples_per_second": 27.891,
+      "eval_steps_per_second": 6.99,
+      "eval_wer": 0.4796796557494621,
+      "step": 1806
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 7.645702838897705,
+      "learning_rate": 9.383003952569171e-05,
+      "loss": 0.2086,
+      "step": 2064
+    },
+    {
+      "epoch": 8.0,
+      "eval_cer": 0.09632947656658067,
+      "eval_loss": 0.4218059182167053,
+      "eval_runtime": 29.7338,
+      "eval_samples_per_second": 27.78,
+      "eval_steps_per_second": 6.962,
+      "eval_wer": 0.4603155629930672,
+      "step": 2064
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 25800,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 100,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "EarlyStoppingCallback": {
+      "args": {
+        "early_stopping_patience": 20,
+        "early_stopping_threshold": 0.001
+      },
+      "attributes": {
+        "early_stopping_patience_counter": 0
+      }
+    },
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 4.3664080259736617e+18,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-2064/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:badc6886eb2995443a897620ab2c36decc3b66898e976799a420ad7fb69a195c
+size 5304