End of training

wav2vec-vlbg-22-attnDrp0.5/config.json

@@ -0,0 +1,80 @@
+{
+  "_name_or_path": "facebook/wav2vec2-base-960h",
+  "activation_dropout": 0.1,
+  "apply_spec_augment": true,
+  "architectures": [
+    "Wav2Vec2ForCTC"
+  ],
+  "attention_dropout": 0.5,
+  "bos_token_id": 1,
+  "classifier_proj_size": 256,
+  "codevector_dim": 256,
+  "contrastive_logits_temperature": 0.1,
+  "conv_bias": false,
+  "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": false,
+  "eos_token_id": 2,
+  "feat_extract_activation": "gelu",
+  "feat_extract_dropout": 0.0,
+  "feat_extract_norm": "group",
+  "feat_proj_dropout": 0.0,
+  "feat_quantizer_dropout": 0.0,
+  "final_dropout": 0.1,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout": 0.0,
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-05,
+  "layerdrop": 0.0,
+  "mask_feature_length": 10,
+  "mask_feature_prob": 0.0,
+  "mask_time_length": 10,
+  "mask_time_prob": 0.05,
+  "model_type": "wav2vec2",
+  "num_attention_heads": 12,
+  "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": 12,
+  "num_negatives": 100,
+  "pad_token_id": 25,
+  "proj_codevector_dim": 256,
+  "torch_dtype": "float32",
+  "transformers_version": "4.11.3",
+  "use_weighted_layer_sum": false,
+  "vocab_size": 32
+}

wav2vec-vlbg-22-attnDrp0.5/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7652bd28d582530a7095eff176fc3de9aed7ff948a28fad6232845e814085ded
+size 721685265

wav2vec-vlbg-22-attnDrp0.5/preprocessor_config.json

@@ -0,0 +1,9 @@
+{
+  "do_normalize": true,
+  "feature_extractor_type": "Wav2Vec2FeatureExtractor",
+  "feature_size": 1,
+  "padding_side": "right",
+  "padding_value": 0.0,
+  "return_attention_mask": true,
+  "sampling_rate": 16000
+}

wav2vec-vlbg-22-attnDrp0.5/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a97d1dc00cc2ffa5a3c94ccc4d856588788e09a1b3162820a3072e7a48c366da
+size 377670039

wav2vec-vlbg-22-attnDrp0.5/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f31b5642254324a4847709ed94b83991885d4075ed00999d0d2889eabb039a9c
+size 14503

wav2vec-vlbg-22-attnDrp0.5/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:695e1b41d52db009945448511c388f551bb25e25a2b7be73b80d70ca69cfda09
+size 559

wav2vec-vlbg-22-attnDrp0.5/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e9825e1cf53d4ea7bf8cc5f370fc01fdccdf6852728c6f671812b644b5fe960e
+size 623

wav2vec-vlbg-22-attnDrp0.5/trainer_state.json

@@ -0,0 +1,256 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 114.27586206896552,
+  "global_step": 1600,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 7.14,
+      "learning_rate": 5.82e-05,
+      "loss": 5.3084,
+      "step": 100
+    },
+    {
+      "epoch": 7.14,
+      "eval_loss": 3.511751651763916,
+      "eval_runtime": 100.7179,
+      "eval_samples_per_second": 1.092,
+      "eval_steps_per_second": 0.139,
+      "eval_wer": 1.0,
+      "step": 100
+    },
+    {
+      "epoch": 14.28,
+      "learning_rate": 0.0001182,
+      "loss": 2.9447,
+      "step": 200
+    },
+    {
+      "epoch": 14.28,
+      "eval_loss": 2.3098528385162354,
+      "eval_runtime": 3.4364,
+      "eval_samples_per_second": 32.01,
+      "eval_steps_per_second": 4.074,
+      "eval_wer": 0.9966722129783694,
+      "step": 200
+    },
+    {
+      "epoch": 21.41,
+      "learning_rate": 0.00017819999999999997,
+      "loss": 1.6778,
+      "step": 300
+    },
+    {
+      "epoch": 21.41,
+      "eval_loss": 1.2265669107437134,
+      "eval_runtime": 3.454,
+      "eval_samples_per_second": 31.847,
+      "eval_steps_per_second": 4.053,
+      "eval_wer": 0.7004991680532446,
+      "step": 300
+    },
+    {
+      "epoch": 28.55,
+      "learning_rate": 0.0002382,
+      "loss": 0.6555,
+      "step": 400
+    },
+    {
+      "epoch": 28.55,
+      "eval_loss": 0.379589319229126,
+      "eval_runtime": 3.4986,
+      "eval_samples_per_second": 31.441,
+      "eval_steps_per_second": 4.002,
+      "eval_wer": 0.3144758735440932,
+      "step": 400
+    },
+    {
+      "epoch": 35.69,
+      "learning_rate": 0.00029759999999999997,
+      "loss": 0.379,
+      "step": 500
+    },
+    {
+      "epoch": 35.69,
+      "eval_loss": 0.38185402750968933,
+      "eval_runtime": 3.4841,
+      "eval_samples_per_second": 31.572,
+      "eval_steps_per_second": 4.018,
+      "eval_wer": 0.281198003327787,
+      "step": 500
+    },
+    {
+      "epoch": 42.83,
+      "learning_rate": 0.000274054054054054,
+      "loss": 0.2896,
+      "step": 600
+    },
+    {
+      "epoch": 42.83,
+      "eval_loss": 0.3479519784450531,
+      "eval_runtime": 3.4717,
+      "eval_samples_per_second": 31.685,
+      "eval_steps_per_second": 4.033,
+      "eval_wer": 0.2096505823627288,
+      "step": 600
+    },
+    {
+      "epoch": 49.97,
+      "learning_rate": 0.000247027027027027,
+      "loss": 0.2347,
+      "step": 700
+    },
+    {
+      "epoch": 49.97,
+      "eval_loss": 0.34583425521850586,
+      "eval_runtime": 3.5221,
+      "eval_samples_per_second": 31.232,
+      "eval_steps_per_second": 3.975,
+      "eval_wer": 0.15640599001663893,
+      "step": 700
+    },
+    {
+      "epoch": 57.14,
+      "learning_rate": 0.00021999999999999995,
+      "loss": 0.1723,
+      "step": 800
+    },
+    {
+      "epoch": 57.14,
+      "eval_loss": 0.2733493149280548,
+      "eval_runtime": 3.5222,
+      "eval_samples_per_second": 31.231,
+      "eval_steps_per_second": 3.975,
+      "eval_wer": 0.13810316139767054,
+      "step": 800
+    },
+    {
+      "epoch": 64.28,
+      "learning_rate": 0.00019297297297297294,
+      "loss": 0.1246,
+      "step": 900
+    },
+    {
+      "epoch": 64.28,
+      "eval_loss": 0.23303170502185822,
+      "eval_runtime": 3.5107,
+      "eval_samples_per_second": 31.333,
+      "eval_steps_per_second": 3.988,
+      "eval_wer": 0.10149750415973377,
+      "step": 900
+    },
+    {
+      "epoch": 71.41,
+      "learning_rate": 0.00016594594594594595,
+      "loss": 0.0996,
+      "step": 1000
+    },
+    {
+      "epoch": 71.41,
+      "eval_loss": 0.279181569814682,
+      "eval_runtime": 3.4803,
+      "eval_samples_per_second": 31.606,
+      "eval_steps_per_second": 4.023,
+      "eval_wer": 0.1064891846921797,
+      "step": 1000
+    },
+    {
+      "epoch": 78.55,
+      "learning_rate": 0.0001389189189189189,
+      "loss": 0.0833,
+      "step": 1100
+    },
+    {
+      "epoch": 78.55,
+      "eval_loss": 0.2734523415565491,
+      "eval_runtime": 3.4763,
+      "eval_samples_per_second": 31.643,
+      "eval_steps_per_second": 4.027,
+      "eval_wer": 0.1048252911813644,
+      "step": 1100
+    },
+    {
+      "epoch": 85.69,
+      "learning_rate": 0.00011189189189189189,
+      "loss": 0.0692,
+      "step": 1200
+    },
+    {
+      "epoch": 85.69,
+      "eval_loss": 0.26076647639274597,
+      "eval_runtime": 3.5052,
+      "eval_samples_per_second": 31.382,
+      "eval_steps_per_second": 3.994,
+      "eval_wer": 0.09151414309484193,
+      "step": 1200
+    },
+    {
+      "epoch": 92.83,
+      "learning_rate": 8.486486486486486e-05,
+      "loss": 0.0475,
+      "step": 1300
+    },
+    {
+      "epoch": 92.83,
+      "eval_loss": 0.29928964376449585,
+      "eval_runtime": 3.4946,
+      "eval_samples_per_second": 31.477,
+      "eval_steps_per_second": 4.006,
+      "eval_wer": 0.08818635607321132,
+      "step": 1300
+    },
+    {
+      "epoch": 99.97,
+      "learning_rate": 5.7837837837837834e-05,
+      "loss": 0.0407,
+      "step": 1400
+    },
+    {
+      "epoch": 99.97,
+      "eval_loss": 0.2704680562019348,
+      "eval_runtime": 3.4967,
+      "eval_samples_per_second": 31.458,
+      "eval_steps_per_second": 4.004,
+      "eval_wer": 0.08153078202995008,
+      "step": 1400
+    },
+    {
+      "epoch": 107.14,
+      "learning_rate": 3.081081081081081e-05,
+      "loss": 0.0345,
+      "step": 1500
+    },
+    {
+      "epoch": 107.14,
+      "eval_loss": 0.30478307604789734,
+      "eval_runtime": 3.5143,
+      "eval_samples_per_second": 31.301,
+      "eval_steps_per_second": 3.984,
+      "eval_wer": 0.07154742096505824,
+      "step": 1500
+    },
+    {
+      "epoch": 114.28,
+      "learning_rate": 3.783783783783783e-06,
+      "loss": 0.0306,
+      "step": 1600
+    },
+    {
+      "epoch": 114.28,
+      "eval_loss": 0.2908855676651001,
+      "eval_runtime": 3.4845,
+      "eval_samples_per_second": 31.568,
+      "eval_steps_per_second": 4.018,
+      "eval_wer": 0.07487520798668885,
+      "step": 1600
+    }
+  ],
+  "max_steps": 1610,
+  "num_train_epochs": 115,
+  "total_flos": 1.3612264016020646e+18,
+  "trial_name": null,
+  "trial_params": null
+}

wav2vec-vlbg-22-attnDrp0.5/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4bc33e3e54fd996c3a8dcffec7bba8308f857e7dc317b86007b6960ed12e3e33
+size 2799