End of training

runs/{May14_20-20-41_f8d43c95907a → xlsrDE-vlbg-L}/events.out.tfevents.1652559651.f8d43c95907a.3830.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0a01a7df8d830e02f78699d305aac2139fd01b314ee59bee5115c82c73a61d66
-size 11927
+oid sha256:c75ec8e8a4b583791922161a171a4156fcd59e4deea6b789eec92df2440399a0
+size 12281

xlsrDE-vlbg-L/config.json

@@ -0,0 +1,86 @@
+{
+  "_name_or_path": "jonatasgrosman/wav2vec2-large-xlsr-53-german",
+  "activation_dropout": 0.05,
+  "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": true,
+  "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,
+  "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_channel_length": 10,
+  "mask_channel_min_space": 1,
+  "mask_channel_other": 0.0,
+  "mask_channel_prob": 0.0,
+  "mask_channel_selection": "static",
+  "mask_feature_length": 10,
+  "mask_feature_prob": 0.0,
+  "mask_time_length": 10,
+  "mask_time_min_space": 1,
+  "mask_time_other": 0.0,
+  "mask_time_prob": 0.05,
+  "mask_time_selection": "static",
+  "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": 38
+}

xlsrDE-vlbg-L/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0db648708f2fd0b480609d2252318245ec00e4f353ac8df989708b5119d60578
+size 2490370577

xlsrDE-vlbg-L/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
+}

xlsrDE-vlbg-L/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:042e4e8b56fc83f3bfa9cff9f2c864dde1a14e6a23b115286047feadc09ef616
+size 1262079473

xlsrDE-vlbg-L/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9285ea9d5d60b2a359908d721f0c25e1a6d9d7ae66ac969f01cefc9cc7a711d7
+size 14503

xlsrDE-vlbg-L/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:91683da6dd64923b7e1df8020e96a40e13b37f7362ff069ffd02a8427bd8db5b
+size 559

xlsrDE-vlbg-L/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f8fd6b64173700554787bd96b3a97002f204277035a984524a8a564f4da1aa28
+size 623

xlsrDE-vlbg-L/trainer_state.json

@@ -0,0 +1,256 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 320.0,
+  "global_step": 1600,
+  "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": 6.8552,
+      "step": 100
+    },
+    {
+      "epoch": 20.0,
+      "eval_loss": 3.0998222827911377,
+      "eval_runtime": 2.8119,
+      "eval_samples_per_second": 21.693,
+      "eval_steps_per_second": 2.845,
+      "eval_wer": 1.0,
+      "step": 100
+    },
+    {
+      "epoch": 40.0,
+      "learning_rate": 0.0001176,
+      "loss": 2.7154,
+      "step": 200
+    },
+    {
+      "epoch": 40.0,
+      "eval_loss": 3.1359784603118896,
+      "eval_runtime": 2.7948,
+      "eval_samples_per_second": 21.826,
+      "eval_steps_per_second": 2.862,
+      "eval_wer": 1.0028818443804035,
+      "step": 200
+    },
+    {
+      "epoch": 60.0,
+      "learning_rate": 0.00017699999999999997,
+      "loss": 1.778,
+      "step": 300
+    },
+    {
+      "epoch": 60.0,
+      "eval_loss": 3.9075214862823486,
+      "eval_runtime": 2.7761,
+      "eval_samples_per_second": 21.973,
+      "eval_steps_per_second": 2.882,
+      "eval_wer": 1.0172910662824208,
+      "step": 300
+    },
+    {
+      "epoch": 80.0,
+      "learning_rate": 0.000237,
+      "loss": 0.8588,
+      "step": 400
+    },
+    {
+      "epoch": 80.0,
+      "eval_loss": 5.115807056427002,
+      "eval_runtime": 2.7812,
+      "eval_samples_per_second": 21.933,
+      "eval_steps_per_second": 2.876,
+      "eval_wer": 1.1354466858789625,
+      "step": 400
+    },
+    {
+      "epoch": 100.0,
+      "learning_rate": 0.00029699999999999996,
+      "loss": 0.4147,
+      "step": 500
+    },
+    {
+      "epoch": 100.0,
+      "eval_loss": 6.190507888793945,
+      "eval_runtime": 2.7947,
+      "eval_samples_per_second": 21.827,
+      "eval_steps_per_second": 2.863,
+      "eval_wer": 1.0403458213256485,
+      "step": 500
+    },
+    {
+      "epoch": 120.0,
+      "learning_rate": 0.0002740909090909091,
+      "loss": 0.21,
+      "step": 600
+    },
+    {
+      "epoch": 120.0,
+      "eval_loss": 6.731902599334717,
+      "eval_runtime": 2.7781,
+      "eval_samples_per_second": 21.957,
+      "eval_steps_per_second": 2.88,
+      "eval_wer": 1.302593659942363,
+      "step": 600
+    },
+    {
+      "epoch": 140.0,
+      "learning_rate": 0.0002468181818181818,
+      "loss": 0.1141,
+      "step": 700
+    },
+    {
+      "epoch": 140.0,
+      "eval_loss": 6.874683380126953,
+      "eval_runtime": 2.8012,
+      "eval_samples_per_second": 21.776,
+      "eval_steps_per_second": 2.856,
+      "eval_wer": 1.1556195965417868,
+      "step": 700
+    },
+    {
+      "epoch": 160.0,
+      "learning_rate": 0.00021954545454545452,
+      "loss": 0.0713,
+      "step": 800
+    },
+    {
+      "epoch": 160.0,
+      "eval_loss": 7.406906604766846,
+      "eval_runtime": 2.7841,
+      "eval_samples_per_second": 21.91,
+      "eval_steps_per_second": 2.873,
+      "eval_wer": 1.2910662824207493,
+      "step": 800
+    },
+    {
+      "epoch": 180.0,
+      "learning_rate": 0.00019227272727272723,
+      "loss": 0.0544,
+      "step": 900
+    },
+    {
+      "epoch": 180.0,
+      "eval_loss": 7.207446575164795,
+      "eval_runtime": 2.7643,
+      "eval_samples_per_second": 22.067,
+      "eval_steps_per_second": 2.894,
+      "eval_wer": 1.1729106628242074,
+      "step": 900
+    },
+    {
+      "epoch": 200.0,
+      "learning_rate": 0.000165,
+      "loss": 0.0427,
+      "step": 1000
+    },
+    {
+      "epoch": 200.0,
+      "eval_loss": 7.732298851013184,
+      "eval_runtime": 2.7743,
+      "eval_samples_per_second": 21.988,
+      "eval_steps_per_second": 2.884,
+      "eval_wer": 1.2219020172910662,
+      "step": 1000
+    },
+    {
+      "epoch": 220.0,
+      "learning_rate": 0.0001377272727272727,
+      "loss": 0.0325,
+      "step": 1100
+    },
+    {
+      "epoch": 220.0,
+      "eval_loss": 8.044499397277832,
+      "eval_runtime": 2.7759,
+      "eval_samples_per_second": 21.975,
+      "eval_steps_per_second": 2.882,
+      "eval_wer": 1.1354466858789625,
+      "step": 1100
+    },
+    {
+      "epoch": 240.0,
+      "learning_rate": 0.00011045454545454545,
+      "loss": 0.0257,
+      "step": 1200
+    },
+    {
+      "epoch": 240.0,
+      "eval_loss": 7.880440711975098,
+      "eval_runtime": 2.7728,
+      "eval_samples_per_second": 21.999,
+      "eval_steps_per_second": 2.885,
+      "eval_wer": 1.1210374639769451,
+      "step": 1200
+    },
+    {
+      "epoch": 260.0,
+      "learning_rate": 8.318181818181818e-05,
+      "loss": 0.0194,
+      "step": 1300
+    },
+    {
+      "epoch": 260.0,
+      "eval_loss": 8.158945083618164,
+      "eval_runtime": 2.779,
+      "eval_samples_per_second": 21.951,
+      "eval_steps_per_second": 2.879,
+      "eval_wer": 1.1469740634005763,
+      "step": 1300
+    },
+    {
+      "epoch": 280.0,
+      "learning_rate": 5.590909090909091e-05,
+      "loss": 0.0164,
+      "step": 1400
+    },
+    {
+      "epoch": 280.0,
+      "eval_loss": 8.227437973022461,
+      "eval_runtime": 2.789,
+      "eval_samples_per_second": 21.872,
+      "eval_steps_per_second": 2.868,
+      "eval_wer": 1.1469740634005763,
+      "step": 1400
+    },
+    {
+      "epoch": 300.0,
+      "learning_rate": 2.8636363636363634e-05,
+      "loss": 0.0132,
+      "step": 1500
+    },
+    {
+      "epoch": 300.0,
+      "eval_loss": 8.220739364624023,
+      "eval_runtime": 2.7747,
+      "eval_samples_per_second": 21.984,
+      "eval_steps_per_second": 2.883,
+      "eval_wer": 1.1527377521613833,
+      "step": 1500
+    },
+    {
+      "epoch": 320.0,
+      "learning_rate": 1.3636363636363634e-06,
+      "loss": 0.0124,
+      "step": 1600
+    },
+    {
+      "epoch": 320.0,
+      "eval_loss": 8.24325942993164,
+      "eval_runtime": 2.7941,
+      "eval_samples_per_second": 21.832,
+      "eval_steps_per_second": 2.863,
+      "eval_wer": 1.2103746397694524,
+      "step": 1600
+    }
+  ],
+  "max_steps": 1600,
+  "num_train_epochs": 320,
+  "total_flos": 6.651185488801571e+18,
+  "trial_name": null,
+  "trial_params": null
+}

xlsrDE-vlbg-L/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f1d669c8f64a0cffec040e94a5f8209add66885af8fd6b962d7634f1c6d12bc
+size 2799