End of training

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6093e8515e44b516179f6ec933a23f5626b5598f55c06137f6d6d19bf788f229
+oid sha256:15700feb753456df987644e3c6cab030166a30211c6e3f506565009e1178918e
 size 1262042545

runs/{May08_00-30-44_dd9a96f88304 → xlsr-vlbg-22DL-attnDrp0.0}/events.out.tfevents.1651969857.dd9a96f88304.3221.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5b7bbdb15a0532dc2218c5fa5ed3ab9855b27ef80a34de9b135330bf3bcf73dc
-size 11740
+oid sha256:e0d6302252284ee905a259de6ac155bb7a8440c7d89a0e02fae96ed393064711
+size 12094

xlsr-vlbg-22DL-attnDrp0.0/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": 0.0,
+  "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-22DL-attnDrp0.0/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0943308b7018df1f190effe4ca9cfacc3d0cd491f288274ff528b891d39f959d
+size 2490296721

xlsr-vlbg-22DL-attnDrp0.0/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-22DL-attnDrp0.0/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6093e8515e44b516179f6ec933a23f5626b5598f55c06137f6d6d19bf788f229
+size 1262042545

xlsr-vlbg-22DL-attnDrp0.0/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b8cb30125368b0b942dfb51d439ea0dbafee5f99a0661db0f31b5bbdc8b5d42a
+size 14567

xlsr-vlbg-22DL-attnDrp0.0/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:26a3b5d4773071e00b720348117bc8f0a7a7319d191fb60cfa69cb50a85a1648
+size 559

xlsr-vlbg-22DL-attnDrp0.0/scheduler.pt

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

xlsr-vlbg-22DL-attnDrp0.0/trainer_state.json

@@ -0,0 +1,256 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 57.14035087719298,
+  "global_step": 1600,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 3.56,
+      "learning_rate": 5.82e-05,
+      "loss": 10.2457,
+      "step": 100
+    },
+    {
+      "epoch": 3.56,
+      "eval_loss": 3.89790678024292,
+      "eval_runtime": 10.8208,
+      "eval_samples_per_second": 20.978,
+      "eval_steps_per_second": 2.68,
+      "eval_wer": 1.0,
+      "step": 100
+    },
+    {
+      "epoch": 7.14,
+      "learning_rate": 0.0001182,
+      "loss": 3.3871,
+      "step": 200
+    },
+    {
+      "epoch": 7.14,
+      "eval_loss": 2.992304563522339,
+      "eval_runtime": 10.7837,
+      "eval_samples_per_second": 21.05,
+      "eval_steps_per_second": 2.689,
+      "eval_wer": 1.0,
+      "step": 200
+    },
+    {
+      "epoch": 10.7,
+      "learning_rate": 0.00017819999999999997,
+      "loss": 2.9516,
+      "step": 300
+    },
+    {
+      "epoch": 10.7,
+      "eval_loss": 2.901867628097534,
+      "eval_runtime": 10.7918,
+      "eval_samples_per_second": 21.034,
+      "eval_steps_per_second": 2.687,
+      "eval_wer": 1.0,
+      "step": 300
+    },
+    {
+      "epoch": 14.28,
+      "learning_rate": 0.0002382,
+      "loss": 2.4874,
+      "step": 400
+    },
+    {
+      "epoch": 14.28,
+      "eval_loss": 1.7228258848190308,
+      "eval_runtime": 10.7891,
+      "eval_samples_per_second": 21.04,
+      "eval_steps_per_second": 2.688,
+      "eval_wer": 0.9451518119490695,
+      "step": 400
+    },
+    {
+      "epoch": 17.84,
+      "learning_rate": 0.0002982,
+      "loss": 1.1546,
+      "step": 500
+    },
+    {
+      "epoch": 17.84,
+      "eval_loss": 1.0063705444335938,
+      "eval_runtime": 10.7925,
+      "eval_samples_per_second": 21.033,
+      "eval_steps_per_second": 2.687,
+      "eval_wer": 0.6758080313418218,
+      "step": 500
+    },
+    {
+      "epoch": 21.42,
+      "learning_rate": 0.00027533898305084746,
+      "loss": 0.5481,
+      "step": 600
+    },
+    {
+      "epoch": 21.42,
+      "eval_loss": 0.9379862546920776,
+      "eval_runtime": 10.7718,
+      "eval_samples_per_second": 21.074,
+      "eval_steps_per_second": 2.692,
+      "eval_wer": 0.5827619980411362,
+      "step": 600
+    },
+    {
+      "epoch": 24.98,
+      "learning_rate": 0.00024991525423728815,
+      "loss": 0.3193,
+      "step": 700
+    },
+    {
+      "epoch": 24.98,
+      "eval_loss": 0.9767209887504578,
+      "eval_runtime": 10.7893,
+      "eval_samples_per_second": 21.039,
+      "eval_steps_per_second": 2.688,
+      "eval_wer": 0.5367286973555337,
+      "step": 700
+    },
+    {
+      "epoch": 28.56,
+      "learning_rate": 0.0002244915254237288,
+      "loss": 0.2157,
+      "step": 800
+    },
+    {
+      "epoch": 28.56,
+      "eval_loss": 0.9500876069068909,
+      "eval_runtime": 10.784,
+      "eval_samples_per_second": 21.05,
+      "eval_steps_per_second": 2.689,
+      "eval_wer": 0.4779627815866797,
+      "step": 800
+    },
+    {
+      "epoch": 32.14,
+      "learning_rate": 0.0001990677966101695,
+      "loss": 0.168,
+      "step": 900
+    },
+    {
+      "epoch": 32.14,
+      "eval_loss": 0.9843384623527527,
+      "eval_runtime": 10.8096,
+      "eval_samples_per_second": 21.0,
+      "eval_steps_per_second": 2.683,
+      "eval_wer": 0.47110675808031344,
+      "step": 900
+    },
+    {
+      "epoch": 35.7,
+      "learning_rate": 0.00017364406779661016,
+      "loss": 0.1106,
+      "step": 1000
+    },
+    {
+      "epoch": 35.7,
+      "eval_loss": 1.0451916456222534,
+      "eval_runtime": 10.756,
+      "eval_samples_per_second": 21.105,
+      "eval_steps_per_second": 2.696,
+      "eval_wer": 0.4622918707149853,
+      "step": 1000
+    },
+    {
+      "epoch": 39.28,
+      "learning_rate": 0.00014822033898305082,
+      "loss": 0.0861,
+      "step": 1100
+    },
+    {
+      "epoch": 39.28,
+      "eval_loss": 1.0691289901733398,
+      "eval_runtime": 10.7818,
+      "eval_samples_per_second": 21.054,
+      "eval_steps_per_second": 2.69,
+      "eval_wer": 0.4515181194906954,
+      "step": 1100
+    },
+    {
+      "epoch": 42.84,
+      "learning_rate": 0.0001227966101694915,
+      "loss": 0.0733,
+      "step": 1200
+    },
+    {
+      "epoch": 42.84,
+      "eval_loss": 1.0812684297561646,
+      "eval_runtime": 10.7616,
+      "eval_samples_per_second": 21.094,
+      "eval_steps_per_second": 2.695,
+      "eval_wer": 0.4240940254652302,
+      "step": 1200
+    },
+    {
+      "epoch": 46.42,
+      "learning_rate": 9.73728813559322e-05,
+      "loss": 0.0565,
+      "step": 1300
+    },
+    {
+      "epoch": 46.42,
+      "eval_loss": 1.0310860872268677,
+      "eval_runtime": 10.7365,
+      "eval_samples_per_second": 21.143,
+      "eval_steps_per_second": 2.701,
+      "eval_wer": 0.4485798237022527,
+      "step": 1300
+    },
+    {
+      "epoch": 49.98,
+      "learning_rate": 7.194915254237287e-05,
+      "loss": 0.0443,
+      "step": 1400
+    },
+    {
+      "epoch": 49.98,
+      "eval_loss": 1.047897458076477,
+      "eval_runtime": 10.6803,
+      "eval_samples_per_second": 21.254,
+      "eval_steps_per_second": 2.715,
+      "eval_wer": 0.43290891283055827,
+      "step": 1400
+    },
+    {
+      "epoch": 53.56,
+      "learning_rate": 4.652542372881355e-05,
+      "loss": 0.0352,
+      "step": 1500
+    },
+    {
+      "epoch": 53.56,
+      "eval_loss": 1.0901097059249878,
+      "eval_runtime": 10.7009,
+      "eval_samples_per_second": 21.213,
+      "eval_steps_per_second": 2.71,
+      "eval_wer": 0.42213516160626835,
+      "step": 1500
+    },
+    {
+      "epoch": 57.14,
+      "learning_rate": 2.1101694915254236e-05,
+      "loss": 0.0277,
+      "step": 1600
+    },
+    {
+      "epoch": 57.14,
+      "eval_loss": 1.0588852167129517,
+      "eval_runtime": 10.7114,
+      "eval_samples_per_second": 21.192,
+      "eval_steps_per_second": 2.707,
+      "eval_wer": 0.42213516160626835,
+      "step": 1600
+    }
+  ],
+  "max_steps": 1680,
+  "num_train_epochs": 60,
+  "total_flos": 4.130905589155786e+18,
+  "trial_name": null,
+  "trial_params": null
+}

xlsr-vlbg-22DL-attnDrp0.0/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1c93f17fe9dcbd8622ba707567af0a02567c0d1f9a6d62038a6197bb8587a402
+size 2799