End of training

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:59c37b7719d2cfb4c52c45aa6d28583c1c6556467f9d328ed588db7a550d5cac
+oid sha256:f6808e18897200694bd41f26cf3a31e5b25e89163ddffe009a6ae736f9f54153
 size 1262042545

runs/{May09_07-50-31_184ce95389f2 → xlsr-vlbg-DL-attnDrp0.5}/events.out.tfevents.1652082670.184ce95389f2.1144.2

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ac527254a89cadee2cb77e1449fdd29d1c24afcbb08b052f36d7d8f7d4de2e73
-size 11738
+oid sha256:7301ececc9ec88f219664b4b3170bfb825130500bdf5066ca0d5fc51fb71e2ee
+size 12092

xlsr-vlbg-DL-attnDrp0.5/config.json

@@ -0,0 +1,79 @@
+{
+  "_name_or_path": "facebook/wav2vec2-xls-r-300m",
+  "activation_dropout": 0.0,
+  "apply_spec_augment": true,
+  "architectures": [
+    "Wav2Vec2ForCTC"
+  ],
+  "attention_dropout": 0.5,
+  "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,
+  "gradient_checkpointing": false,
+  "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": 10,
+  "mask_feature_prob": 0.0,
+  "mask_time_length": 10,
+  "mask_time_prob": 0.05,
+  "model_type": "wav2vec2",
+  "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,
+  "pad_token_id": 26,
+  "proj_codevector_dim": 768,
+  "torch_dtype": "float32",
+  "transformers_version": "4.11.3",
+  "use_weighted_layer_sum": false,
+  "vocab_size": 29
+}

xlsr-vlbg-DL-attnDrp0.5/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:46f9064c5b0751b90159f30cf3c118b4badf6eb18a9620926da162a3cae2a1f1
+size 2490296721

xlsr-vlbg-DL-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
+}

xlsr-vlbg-DL-attnDrp0.5/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:59c37b7719d2cfb4c52c45aa6d28583c1c6556467f9d328ed588db7a550d5cac
+size 1262042545

xlsr-vlbg-DL-attnDrp0.5/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d4f9d9f46c6b6c01d380c15b7fd000e9effaad016ae32620c6131e02933759e6
+size 14567

xlsr-vlbg-DL-attnDrp0.5/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1c42decc74b17e4851fcc51766fe3ddc893e4d6f043335e1c4a2764999ef5968
+size 559

xlsr-vlbg-DL-attnDrp0.5/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:767d1694e9ef2cc1caba77a48749c3f99ece3803691fca634506d9ae870315b2
+size 623

xlsr-vlbg-DL-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": 14.2075,
+      "step": 100
+    },
+    {
+      "epoch": 7.14,
+      "eval_loss": 3.9464569091796875,
+      "eval_runtime": 31.1841,
+      "eval_samples_per_second": 3.72,
+      "eval_steps_per_second": 0.481,
+      "eval_wer": 1.0,
+      "step": 100
+    },
+    {
+      "epoch": 14.28,
+      "learning_rate": 0.0001182,
+      "loss": 3.4162,
+      "step": 200
+    },
+    {
+      "epoch": 14.28,
+      "eval_loss": 3.072521686553955,
+      "eval_runtime": 4.3824,
+      "eval_samples_per_second": 26.47,
+      "eval_steps_per_second": 3.423,
+      "eval_wer": 1.0,
+      "step": 200
+    },
+    {
+      "epoch": 21.41,
+      "learning_rate": 0.00017819999999999997,
+      "loss": 2.9811,
+      "step": 300
+    },
+    {
+      "epoch": 21.41,
+      "eval_loss": 2.8972666263580322,
+      "eval_runtime": 4.361,
+      "eval_samples_per_second": 26.599,
+      "eval_steps_per_second": 3.44,
+      "eval_wer": 1.0,
+      "step": 300
+    },
+    {
+      "epoch": 28.55,
+      "learning_rate": 0.0002382,
+      "loss": 2.5037,
+      "step": 400
+    },
+    {
+      "epoch": 28.55,
+      "eval_loss": 2.10002064704895,
+      "eval_runtime": 4.3876,
+      "eval_samples_per_second": 26.438,
+      "eval_steps_per_second": 3.419,
+      "eval_wer": 1.0191846522781776,
+      "step": 400
+    },
+    {
+      "epoch": 35.69,
+      "learning_rate": 0.0002982,
+      "loss": 1.2743,
+      "step": 500
+    },
+    {
+      "epoch": 35.69,
+      "eval_loss": 1.6965256929397583,
+      "eval_runtime": 4.3765,
+      "eval_samples_per_second": 26.505,
+      "eval_steps_per_second": 3.427,
+      "eval_wer": 0.9448441247002398,
+      "step": 500
+    },
+    {
+      "epoch": 42.83,
+      "learning_rate": 0.00027533898305084746,
+      "loss": 0.6392,
+      "step": 600
+    },
+    {
+      "epoch": 42.83,
+      "eval_loss": 1.8076837062835693,
+      "eval_runtime": 4.3432,
+      "eval_samples_per_second": 26.708,
+      "eval_steps_per_second": 3.454,
+      "eval_wer": 0.8657074340527577,
+      "step": 600
+    },
+    {
+      "epoch": 49.97,
+      "learning_rate": 0.00024991525423728815,
+      "loss": 0.3871,
+      "step": 700
+    },
+    {
+      "epoch": 49.97,
+      "eval_loss": 2.100900650024414,
+      "eval_runtime": 4.3648,
+      "eval_samples_per_second": 26.576,
+      "eval_steps_per_second": 3.437,
+      "eval_wer": 0.8585131894484412,
+      "step": 700
+    },
+    {
+      "epoch": 57.14,
+      "learning_rate": 0.0002244915254237288,
+      "loss": 0.2691,
+      "step": 800
+    },
+    {
+      "epoch": 57.14,
+      "eval_loss": 2.047531843185425,
+      "eval_runtime": 4.388,
+      "eval_samples_per_second": 26.436,
+      "eval_steps_per_second": 3.418,
+      "eval_wer": 0.8201438848920863,
+      "step": 800
+    },
+    {
+      "epoch": 64.28,
+      "learning_rate": 0.0001990677966101695,
+      "loss": 0.2032,
+      "step": 900
+    },
+    {
+      "epoch": 64.28,
+      "eval_loss": 2.1936190128326416,
+      "eval_runtime": 4.3587,
+      "eval_samples_per_second": 26.613,
+      "eval_steps_per_second": 3.441,
+      "eval_wer": 0.8537170263788969,
+      "step": 900
+    },
+    {
+      "epoch": 71.41,
+      "learning_rate": 0.00017364406779661016,
+      "loss": 0.1437,
+      "step": 1000
+    },
+    {
+      "epoch": 71.41,
+      "eval_loss": 2.1932051181793213,
+      "eval_runtime": 4.3787,
+      "eval_samples_per_second": 26.492,
+      "eval_steps_per_second": 3.426,
+      "eval_wer": 0.8057553956834532,
+      "step": 1000
+    },
+    {
+      "epoch": 78.55,
+      "learning_rate": 0.00014822033898305082,
+      "loss": 0.1144,
+      "step": 1100
+    },
+    {
+      "epoch": 78.55,
+      "eval_loss": 2.35802960395813,
+      "eval_runtime": 4.3463,
+      "eval_samples_per_second": 26.689,
+      "eval_steps_per_second": 3.451,
+      "eval_wer": 0.8201438848920863,
+      "step": 1100
+    },
+    {
+      "epoch": 85.69,
+      "learning_rate": 0.0001227966101694915,
+      "loss": 0.0938,
+      "step": 1200
+    },
+    {
+      "epoch": 85.69,
+      "eval_loss": 2.427619218826294,
+      "eval_runtime": 4.3556,
+      "eval_samples_per_second": 26.633,
+      "eval_steps_per_second": 3.444,
+      "eval_wer": 0.7937649880095923,
+      "step": 1200
+    },
+    {
+      "epoch": 92.83,
+      "learning_rate": 9.73728813559322e-05,
+      "loss": 0.0672,
+      "step": 1300
+    },
+    {
+      "epoch": 92.83,
+      "eval_loss": 2.7036306858062744,
+      "eval_runtime": 4.3543,
+      "eval_samples_per_second": 26.64,
+      "eval_steps_per_second": 3.445,
+      "eval_wer": 0.829736211031175,
+      "step": 1300
+    },
+    {
+      "epoch": 99.97,
+      "learning_rate": 7.194915254237287e-05,
+      "loss": 0.055,
+      "step": 1400
+    },
+    {
+      "epoch": 99.97,
+      "eval_loss": 2.60349440574646,
+      "eval_runtime": 4.3659,
+      "eval_samples_per_second": 26.57,
+      "eval_steps_per_second": 3.436,
+      "eval_wer": 0.7673860911270983,
+      "step": 1400
+    },
+    {
+      "epoch": 107.14,
+      "learning_rate": 4.652542372881355e-05,
+      "loss": 0.0467,
+      "step": 1500
+    },
+    {
+      "epoch": 107.14,
+      "eval_loss": 2.7448980808258057,
+      "eval_runtime": 4.359,
+      "eval_samples_per_second": 26.611,
+      "eval_steps_per_second": 3.441,
+      "eval_wer": 0.7745803357314148,
+      "step": 1500
+    },
+    {
+      "epoch": 114.28,
+      "learning_rate": 2.1101694915254236e-05,
+      "loss": 0.0372,
+      "step": 1600
+    },
+    {
+      "epoch": 114.28,
+      "eval_loss": 2.6863534450531006,
+      "eval_runtime": 4.341,
+      "eval_samples_per_second": 26.722,
+      "eval_steps_per_second": 3.455,
+      "eval_wer": 0.7985611510791367,
+      "step": 1600
+    }
+  ],
+  "max_steps": 1680,
+  "num_train_epochs": 120,
+  "total_flos": 4.5111935443900457e+18,
+  "trial_name": null,
+  "trial_params": null
+}

xlsr-vlbg-DL-attnDrp0.5/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cd329c80927d4e413ddccff18fcb5454f1b2a52fec4fae399e5e39df1e9a2d1d
+size 2799