End of training

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c8909defc3cfc80e6b475ceee85e815fd923c580f1511cb1643393e515feb106
+oid sha256:f7f8503c006665661c68576dafc534b6b2ee8f4f9cc0e0cba2a3d1efcc05b812
 size 377670039

runs/{May10_17-42-19_726f2b3fea59 → wav2vec-vlbg-MW-attnDrp0.5}/events.out.tfevents.1652204555.726f2b3fea59.25433.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6afdefe4ae5525113014c93d3e20c68488873bc4a5a02db6933b251ede38a1f8
-size 10307
+oid sha256:71db4784ded84d8076c817f26639cb241cfea5c7f923354d454eb49b0ea5c978
+size 13632

wav2vec-vlbg-MW-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": 26,
+  "proj_codevector_dim": 256,
+  "torch_dtype": "float32",
+  "transformers_version": "4.11.3",
+  "use_weighted_layer_sum": false,
+  "vocab_size": 32
+}

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

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:79a2811a89711295998bfb2c18a244ba1bf1c3bc6e8985402a1a1b81249fa75e
+size 721685265

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

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c8909defc3cfc80e6b475ceee85e815fd923c580f1511cb1643393e515feb106
+size 377670039

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

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:389b39bc8670d836abf3b9aac21c1ee40b37ddc4c841788548f2236180c29608
+size 14567

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

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1c33a536167719b6e259a041f22f1cf5636960275aa70b4ce1df2e294934bde6
+size 559

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

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a7f75d9feebadf0d4f1149bfaf72f291ca0d2c3ddba54fa618274654c1a6ab9c
+size 623

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

@@ -0,0 +1,211 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 260.0,
+  "global_step": 1300,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 20.0,
+      "learning_rate": 5.76e-05,
+      "loss": 5.5776,
+      "step": 100
+    },
+    {
+      "epoch": 20.0,
+      "eval_loss": 36.25331497192383,
+      "eval_runtime": 0.6005,
+      "eval_samples_per_second": 43.298,
+      "eval_steps_per_second": 6.661,
+      "eval_wer": 1.0,
+      "step": 100
+    },
+    {
+      "epoch": 40.0,
+      "learning_rate": 0.0001176,
+      "loss": 2.9662,
+      "step": 200
+    },
+    {
+      "epoch": 40.0,
+      "eval_loss": 19.664457321166992,
+      "eval_runtime": 0.5933,
+      "eval_samples_per_second": 43.819,
+      "eval_steps_per_second": 6.741,
+      "eval_wer": 1.0,
+      "step": 200
+    },
+    {
+      "epoch": 60.0,
+      "learning_rate": 0.00017759999999999998,
+      "loss": 2.8096,
+      "step": 300
+    },
+    {
+      "epoch": 60.0,
+      "eval_loss": 41.18692398071289,
+      "eval_runtime": 0.6025,
+      "eval_samples_per_second": 43.153,
+      "eval_steps_per_second": 6.639,
+      "eval_wer": 1.162162162162162,
+      "step": 300
+    },
+    {
+      "epoch": 80.0,
+      "learning_rate": 0.0002376,
+      "loss": 2.4675,
+      "step": 400
+    },
+    {
+      "epoch": 80.0,
+      "eval_loss": 27.60628890991211,
+      "eval_runtime": 0.6028,
+      "eval_samples_per_second": 43.135,
+      "eval_steps_per_second": 6.636,
+      "eval_wer": 1.5135135135135136,
+      "step": 400
+    },
+    {
+      "epoch": 100.0,
+      "learning_rate": 0.00029759999999999997,
+      "loss": 1.7554,
+      "step": 500
+    },
+    {
+      "epoch": 100.0,
+      "eval_loss": 40.952999114990234,
+      "eval_runtime": 0.5943,
+      "eval_samples_per_second": 43.749,
+      "eval_steps_per_second": 6.731,
+      "eval_wer": 1.7027027027027026,
+      "step": 500
+    },
+    {
+      "epoch": 120.0,
+      "learning_rate": 0.0002869090909090909,
+      "loss": 1.0254,
+      "step": 600
+    },
+    {
+      "epoch": 120.0,
+      "eval_loss": 32.34526824951172,
+      "eval_runtime": 0.5931,
+      "eval_samples_per_second": 43.834,
+      "eval_steps_per_second": 6.744,
+      "eval_wer": 1.8108108108108107,
+      "step": 600
+    },
+    {
+      "epoch": 140.0,
+      "learning_rate": 0.00027327272727272725,
+      "loss": 0.6174,
+      "step": 700
+    },
+    {
+      "epoch": 140.0,
+      "eval_loss": 34.84318923950195,
+      "eval_runtime": 0.5959,
+      "eval_samples_per_second": 43.629,
+      "eval_steps_per_second": 6.712,
+      "eval_wer": 1.864864864864865,
+      "step": 700
+    },
+    {
+      "epoch": 160.0,
+      "learning_rate": 0.00025963636363636363,
+      "loss": 0.4281,
+      "step": 800
+    },
+    {
+      "epoch": 160.0,
+      "eval_loss": 36.15171813964844,
+      "eval_runtime": 0.5965,
+      "eval_samples_per_second": 43.588,
+      "eval_steps_per_second": 6.706,
+      "eval_wer": 1.864864864864865,
+      "step": 800
+    },
+    {
+      "epoch": 180.0,
+      "learning_rate": 0.00024613636363636363,
+      "loss": 0.324,
+      "step": 900
+    },
+    {
+      "epoch": 180.0,
+      "eval_loss": 37.320735931396484,
+      "eval_runtime": 0.5968,
+      "eval_samples_per_second": 43.567,
+      "eval_steps_per_second": 6.703,
+      "eval_wer": 1.8918918918918919,
+      "step": 900
+    },
+    {
+      "epoch": 200.0,
+      "learning_rate": 0.00023249999999999999,
+      "loss": 0.2476,
+      "step": 1000
+    },
+    {
+      "epoch": 200.0,
+      "eval_loss": 41.81070327758789,
+      "eval_runtime": 0.5943,
+      "eval_samples_per_second": 43.747,
+      "eval_steps_per_second": 6.73,
+      "eval_wer": 1.864864864864865,
+      "step": 1000
+    },
+    {
+      "epoch": 220.0,
+      "learning_rate": 0.00021899999999999998,
+      "loss": 0.2166,
+      "step": 1100
+    },
+    {
+      "epoch": 220.0,
+      "eval_loss": 39.16055679321289,
+      "eval_runtime": 0.5903,
+      "eval_samples_per_second": 44.046,
+      "eval_steps_per_second": 6.776,
+      "eval_wer": 1.8918918918918919,
+      "step": 1100
+    },
+    {
+      "epoch": 240.0,
+      "learning_rate": 0.00020536363636363637,
+      "loss": 0.1803,
+      "step": 1200
+    },
+    {
+      "epoch": 240.0,
+      "eval_loss": 41.908931732177734,
+      "eval_runtime": 0.5698,
+      "eval_samples_per_second": 45.631,
+      "eval_steps_per_second": 7.02,
+      "eval_wer": 1.8918918918918919,
+      "step": 1200
+    },
+    {
+      "epoch": 260.0,
+      "learning_rate": 0.0001917272727272727,
+      "loss": 0.1429,
+      "step": 1300
+    },
+    {
+      "epoch": 260.0,
+      "eval_loss": 39.956539154052734,
+      "eval_runtime": 0.592,
+      "eval_samples_per_second": 43.917,
+      "eval_steps_per_second": 6.756,
+      "eval_wer": 1.837837837837838,
+      "step": 1300
+    }
+  ],
+  "max_steps": 2700,
+  "num_train_epochs": 540,
+  "total_flos": 1.6132333708441843e+18,
+  "trial_name": null,
+  "trial_params": null
+}

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

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:682fecbdd7eb89fec95b3ed1a209e233fa42412664c99cdc38e76cd95320a6d3
+size 2799