End of training

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5690a0081b090144361236f48250335e6858018fc61f945789c309c53d2c5ed1
+oid sha256:17f4b7b632aa7c1fb8dcfd3595da9dc8d4a1c4a9d6b5f28a76940fc9205eed53
 size 1262038449

runs/{May09_10-10-31_184ce95389f2 → xlsr-vlbg-22D-attnDrp0.5}/events.out.tfevents.1652091093.184ce95389f2.4845.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3478c605c9b10d6c62e50493aed4b0ea81d15ac58da60d2a0bcce752874a81c6
-size 11737
+oid sha256:20e00333c86913b6f58f8878cc40031258c2d4a7df43287584ea1a288b4ea3ab
+size 12091

xlsr-vlbg-22D-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": 25,
+  "proj_codevector_dim": 768,
+  "torch_dtype": "float32",
+  "transformers_version": "4.11.3",
+  "use_weighted_layer_sum": false,
+  "vocab_size": 28
+}

xlsr-vlbg-22D-attnDrp0.5/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:65a23f404b9b2b4648f4ed5e203d57c01faaf1da7e323cc70e82853212e4fa45
+size 2490288529

xlsr-vlbg-22D-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-22D-attnDrp0.5/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5690a0081b090144361236f48250335e6858018fc61f945789c309c53d2c5ed1
+size 1262038449

xlsr-vlbg-22D-attnDrp0.5/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2aaf332cfc44175f888e49e58a020b990d3f7e192023353b6b117ad183650e08
+size 14567

xlsr-vlbg-22D-attnDrp0.5/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc7dd9f51a85ba3bb1a1c29e8c0d4cddc6462654d83011d6ca2b11e1865a1062
+size 559

xlsr-vlbg-22D-attnDrp0.5/scheduler.pt

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

xlsr-vlbg-22D-attnDrp0.5/trainer_state.json

@@ -0,0 +1,256 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 69.55319148936171,
+  "global_step": 1600,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 4.34,
+      "learning_rate": 5.76e-05,
+      "loss": 11.0952,
+      "step": 100
+    },
+    {
+      "epoch": 4.34,
+      "eval_loss": 3.9931130409240723,
+      "eval_runtime": 40.9643,
+      "eval_samples_per_second": 4.028,
+      "eval_steps_per_second": 0.513,
+      "eval_wer": 1.0,
+      "step": 100
+    },
+    {
+      "epoch": 8.68,
+      "learning_rate": 0.0001176,
+      "loss": 3.3935,
+      "step": 200
+    },
+    {
+      "epoch": 8.68,
+      "eval_loss": 3.0509140491485596,
+      "eval_runtime": 6.4437,
+      "eval_samples_per_second": 25.606,
+      "eval_steps_per_second": 3.259,
+      "eval_wer": 1.0,
+      "step": 200
+    },
+    {
+      "epoch": 13.04,
+      "learning_rate": 0.00017759999999999998,
+      "loss": 2.9376,
+      "step": 300
+    },
+    {
+      "epoch": 13.04,
+      "eval_loss": 2.5794737339019775,
+      "eval_runtime": 6.4626,
+      "eval_samples_per_second": 25.532,
+      "eval_steps_per_second": 3.249,
+      "eval_wer": 1.0,
+      "step": 300
+    },
+    {
+      "epoch": 17.38,
+      "learning_rate": 0.0002376,
+      "loss": 1.9559,
+      "step": 400
+    },
+    {
+      "epoch": 17.38,
+      "eval_loss": 1.1511459350585938,
+      "eval_runtime": 6.4461,
+      "eval_samples_per_second": 25.597,
+      "eval_steps_per_second": 3.258,
+      "eval_wer": 0.7555886736214605,
+      "step": 400
+    },
+    {
+      "epoch": 21.72,
+      "learning_rate": 0.00029759999999999997,
+      "loss": 0.7716,
+      "step": 500
+    },
+    {
+      "epoch": 21.72,
+      "eval_loss": 1.0048249959945679,
+      "eval_runtime": 6.428,
+      "eval_samples_per_second": 25.669,
+      "eval_steps_per_second": 3.267,
+      "eval_wer": 0.44113263785394935,
+      "step": 500
+    },
+    {
+      "epoch": 26.09,
+      "learning_rate": 0.000274054054054054,
+      "loss": 0.4573,
+      "step": 600
+    },
+    {
+      "epoch": 26.09,
+      "eval_loss": 0.8895505666732788,
+      "eval_runtime": 6.4489,
+      "eval_samples_per_second": 25.586,
+      "eval_steps_per_second": 3.256,
+      "eval_wer": 0.37257824143070045,
+      "step": 600
+    },
+    {
+      "epoch": 30.43,
+      "learning_rate": 0.000247027027027027,
+      "loss": 0.2741,
+      "step": 700
+    },
+    {
+      "epoch": 30.43,
+      "eval_loss": 0.857887327671051,
+      "eval_runtime": 6.4667,
+      "eval_samples_per_second": 25.515,
+      "eval_steps_per_second": 3.247,
+      "eval_wer": 0.30402384500745155,
+      "step": 700
+    },
+    {
+      "epoch": 34.77,
+      "learning_rate": 0.00021999999999999995,
+      "loss": 0.1894,
+      "step": 800
+    },
+    {
+      "epoch": 34.77,
+      "eval_loss": 0.8584926724433899,
+      "eval_runtime": 6.4574,
+      "eval_samples_per_second": 25.552,
+      "eval_steps_per_second": 3.252,
+      "eval_wer": 0.27123695976154993,
+      "step": 800
+    },
+    {
+      "epoch": 39.13,
+      "learning_rate": 0.00019297297297297294,
+      "loss": 0.1446,
+      "step": 900
+    },
+    {
+      "epoch": 39.13,
+      "eval_loss": 0.9569060206413269,
+      "eval_runtime": 6.4887,
+      "eval_samples_per_second": 25.429,
+      "eval_steps_per_second": 3.236,
+      "eval_wer": 0.23845007451564829,
+      "step": 900
+    },
+    {
+      "epoch": 43.47,
+      "learning_rate": 0.00016594594594594595,
+      "loss": 0.1108,
+      "step": 1000
+    },
+    {
+      "epoch": 43.47,
+      "eval_loss": 1.0971928834915161,
+      "eval_runtime": 6.4847,
+      "eval_samples_per_second": 25.445,
+      "eval_steps_per_second": 3.238,
+      "eval_wer": 0.2637853949329359,
+      "step": 1000
+    },
+    {
+      "epoch": 47.81,
+      "learning_rate": 0.0001389189189189189,
+      "loss": 0.0844,
+      "step": 1100
+    },
+    {
+      "epoch": 47.81,
+      "eval_loss": 0.9556686878204346,
+      "eval_runtime": 6.4425,
+      "eval_samples_per_second": 25.611,
+      "eval_steps_per_second": 3.26,
+      "eval_wer": 0.27421758569299554,
+      "step": 1100
+    },
+    {
+      "epoch": 52.17,
+      "learning_rate": 0.00011189189189189189,
+      "loss": 0.0633,
+      "step": 1200
+    },
+    {
+      "epoch": 52.17,
+      "eval_loss": 1.0082427263259888,
+      "eval_runtime": 6.4179,
+      "eval_samples_per_second": 25.71,
+      "eval_steps_per_second": 3.272,
+      "eval_wer": 0.2503725782414307,
+      "step": 1200
+    },
+    {
+      "epoch": 56.51,
+      "learning_rate": 8.486486486486486e-05,
+      "loss": 0.052,
+      "step": 1300
+    },
+    {
+      "epoch": 56.51,
+      "eval_loss": 1.0009304285049438,
+      "eval_runtime": 6.4498,
+      "eval_samples_per_second": 25.582,
+      "eval_steps_per_second": 3.256,
+      "eval_wer": 0.2563338301043219,
+      "step": 1300
+    },
+    {
+      "epoch": 60.85,
+      "learning_rate": 5.7837837837837834e-05,
+      "loss": 0.0429,
+      "step": 1400
+    },
+    {
+      "epoch": 60.85,
+      "eval_loss": 1.0997564792633057,
+      "eval_runtime": 6.5266,
+      "eval_samples_per_second": 25.281,
+      "eval_steps_per_second": 3.218,
+      "eval_wer": 0.23546944858420268,
+      "step": 1400
+    },
+    {
+      "epoch": 65.21,
+      "learning_rate": 3.081081081081081e-05,
+      "loss": 0.0373,
+      "step": 1500
+    },
+    {
+      "epoch": 65.21,
+      "eval_loss": 1.0540827512741089,
+      "eval_runtime": 6.4362,
+      "eval_samples_per_second": 25.636,
+      "eval_steps_per_second": 3.263,
+      "eval_wer": 0.2488822652757079,
+      "step": 1500
+    },
+    {
+      "epoch": 69.55,
+      "learning_rate": 3.783783783783783e-06,
+      "loss": 0.0312,
+      "step": 1600
+    },
+    {
+      "epoch": 69.55,
+      "eval_loss": 1.0796761512756348,
+      "eval_runtime": 6.4899,
+      "eval_samples_per_second": 25.424,
+      "eval_steps_per_second": 3.236,
+      "eval_wer": 0.23845007451564829,
+      "step": 1600
+    }
+  ],
+  "max_steps": 1610,
+  "num_train_epochs": 70,
+  "total_flos": 4.2479688746902845e+18,
+  "trial_name": null,
+  "trial_params": null
+}

xlsr-vlbg-22D-attnDrp0.5/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bcfa182d8076519402f16132910be8008c2ca0fda7c7aa40a971d2dc03f01a3d
+size 2799