End of training

runs/{May11_20-23-50_b215901a7a4c → xlsrDE-vlbg-L-attnDrp0.5}/events.out.tfevents.1652300641.b215901a7a4c.15037.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9a6d30bb275764e7856f5cb98d731da58d2007b786e503eb32e6c3aa71c5df44
-size 11927
+oid sha256:ea6eb672bbaf83f6453187fa73454edcfc42e1dffc1a9b57baec901eb6e35628
+size 12281

xlsrDE-vlbg-L-attnDrp0.5/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-attnDrp0.5/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ef7c580ab76cb64b25057eeedaed2c33132d39b998f3bfdc8835b48b965504fa
+size 2490370577

xlsrDE-vlbg-L-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
+}

xlsrDE-vlbg-L-attnDrp0.5/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fbabad655dc16509eedb40ccc85fdef40cd66bda898975944b1e9f87089fbe80
+size 1262079473

xlsrDE-vlbg-L-attnDrp0.5/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dfa6d34e6b8d684c7f9a1075c1c40208dd4b5aa065f1c5ea1a307748166b1d98
+size 14503

xlsrDE-vlbg-L-attnDrp0.5/scaler.pt

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

xlsrDE-vlbg-L-attnDrp0.5/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:67f765ba94b89b46e5e011aedcc3e1309dae4481a46a935100100a450fd8e95f
+size 623

xlsrDE-vlbg-L-attnDrp0.5/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": 5.82e-05,
+      "loss": 6.61,
+      "step": 100
+    },
+    {
+      "epoch": 20.0,
+      "eval_loss": 3.066619396209717,
+      "eval_runtime": 18.029,
+      "eval_samples_per_second": 3.383,
+      "eval_steps_per_second": 0.444,
+      "eval_wer": 1.0,
+      "step": 100
+    },
+    {
+      "epoch": 40.0,
+      "learning_rate": 0.0001182,
+      "loss": 2.6693,
+      "step": 200
+    },
+    {
+      "epoch": 40.0,
+      "eval_loss": 3.044706344604492,
+      "eval_runtime": 2.6038,
+      "eval_samples_per_second": 23.427,
+      "eval_steps_per_second": 3.072,
+      "eval_wer": 0.9970501474926253,
+      "step": 200
+    },
+    {
+      "epoch": 60.0,
+      "learning_rate": 0.00017819999999999997,
+      "loss": 1.7005,
+      "step": 300
+    },
+    {
+      "epoch": 60.0,
+      "eval_loss": 4.028779983520508,
+      "eval_runtime": 2.6,
+      "eval_samples_per_second": 23.461,
+      "eval_steps_per_second": 3.077,
+      "eval_wer": 1.0117994100294985,
+      "step": 300
+    },
+    {
+      "epoch": 80.0,
+      "learning_rate": 0.0002382,
+      "loss": 0.7894,
+      "step": 400
+    },
+    {
+      "epoch": 80.0,
+      "eval_loss": 5.160449981689453,
+      "eval_runtime": 2.6249,
+      "eval_samples_per_second": 23.239,
+      "eval_steps_per_second": 3.048,
+      "eval_wer": 1.0265486725663717,
+      "step": 400
+    },
+    {
+      "epoch": 100.0,
+      "learning_rate": 0.0002982,
+      "loss": 0.3588,
+      "step": 500
+    },
+    {
+      "epoch": 100.0,
+      "eval_loss": 6.262679100036621,
+      "eval_runtime": 2.5874,
+      "eval_samples_per_second": 23.576,
+      "eval_steps_per_second": 3.092,
+      "eval_wer": 1.056047197640118,
+      "step": 500
+    },
+    {
+      "epoch": 120.0,
+      "learning_rate": 0.00027354545454545454,
+      "loss": 0.1847,
+      "step": 600
+    },
+    {
+      "epoch": 120.0,
+      "eval_loss": 6.131147384643555,
+      "eval_runtime": 2.6535,
+      "eval_samples_per_second": 22.988,
+      "eval_steps_per_second": 3.015,
+      "eval_wer": 1.0707964601769913,
+      "step": 600
+    },
+    {
+      "epoch": 140.0,
+      "learning_rate": 0.00024627272727272725,
+      "loss": 0.1079,
+      "step": 700
+    },
+    {
+      "epoch": 140.0,
+      "eval_loss": 6.570610046386719,
+      "eval_runtime": 2.6128,
+      "eval_samples_per_second": 23.346,
+      "eval_steps_per_second": 3.062,
+      "eval_wer": 1.0914454277286136,
+      "step": 700
+    },
+    {
+      "epoch": 160.0,
+      "learning_rate": 0.00021899999999999998,
+      "loss": 0.0723,
+      "step": 800
+    },
+    {
+      "epoch": 160.0,
+      "eval_loss": 7.387508392333984,
+      "eval_runtime": 2.616,
+      "eval_samples_per_second": 23.318,
+      "eval_steps_per_second": 3.058,
+      "eval_wer": 1.1976401179941003,
+      "step": 800
+    },
+    {
+      "epoch": 180.0,
+      "learning_rate": 0.0001917272727272727,
+      "loss": 0.0506,
+      "step": 900
+    },
+    {
+      "epoch": 180.0,
+      "eval_loss": 7.059518337249756,
+      "eval_runtime": 2.617,
+      "eval_samples_per_second": 23.309,
+      "eval_steps_per_second": 3.057,
+      "eval_wer": 1.1740412979351031,
+      "step": 900
+    },
+    {
+      "epoch": 200.0,
+      "learning_rate": 0.00016445454545454546,
+      "loss": 0.0378,
+      "step": 1000
+    },
+    {
+      "epoch": 200.0,
+      "eval_loss": 7.530640125274658,
+      "eval_runtime": 2.63,
+      "eval_samples_per_second": 23.194,
+      "eval_steps_per_second": 3.042,
+      "eval_wer": 1.1445427728613569,
+      "step": 1000
+    },
+    {
+      "epoch": 220.0,
+      "learning_rate": 0.00013718181818181817,
+      "loss": 0.041,
+      "step": 1100
+    },
+    {
+      "epoch": 220.0,
+      "eval_loss": 7.453554153442383,
+      "eval_runtime": 2.6063,
+      "eval_samples_per_second": 23.405,
+      "eval_steps_per_second": 3.07,
+      "eval_wer": 1.0766961651917404,
+      "step": 1100
+    },
+    {
+      "epoch": 240.0,
+      "learning_rate": 0.0001099090909090909,
+      "loss": 0.0254,
+      "step": 1200
+    },
+    {
+      "epoch": 240.0,
+      "eval_loss": 7.780274391174316,
+      "eval_runtime": 2.6442,
+      "eval_samples_per_second": 23.07,
+      "eval_steps_per_second": 3.026,
+      "eval_wer": 1.1504424778761062,
+      "step": 1200
+    },
+    {
+      "epoch": 260.0,
+      "learning_rate": 8.263636363636363e-05,
+      "loss": 0.0211,
+      "step": 1300
+    },
+    {
+      "epoch": 260.0,
+      "eval_loss": 7.915604591369629,
+      "eval_runtime": 2.6109,
+      "eval_samples_per_second": 23.363,
+      "eval_steps_per_second": 3.064,
+      "eval_wer": 1.1002949852507375,
+      "step": 1300
+    },
+    {
+      "epoch": 280.0,
+      "learning_rate": 5.536363636363636e-05,
+      "loss": 0.0175,
+      "step": 1400
+    },
+    {
+      "epoch": 280.0,
+      "eval_loss": 8.009050369262695,
+      "eval_runtime": 2.6199,
+      "eval_samples_per_second": 23.283,
+      "eval_steps_per_second": 3.054,
+      "eval_wer": 1.168141592920354,
+      "step": 1400
+    },
+    {
+      "epoch": 300.0,
+      "learning_rate": 2.809090909090909e-05,
+      "loss": 0.0141,
+      "step": 1500
+    },
+    {
+      "epoch": 300.0,
+      "eval_loss": 8.053727149963379,
+      "eval_runtime": 2.6225,
+      "eval_samples_per_second": 23.26,
+      "eval_steps_per_second": 3.051,
+      "eval_wer": 1.1474926253687316,
+      "step": 1500
+    },
+    {
+      "epoch": 320.0,
+      "learning_rate": 8.181818181818181e-07,
+      "loss": 0.0132,
+      "step": 1600
+    },
+    {
+      "epoch": 320.0,
+      "eval_loss": 8.048476219177246,
+      "eval_runtime": 2.6273,
+      "eval_samples_per_second": 23.218,
+      "eval_steps_per_second": 3.045,
+      "eval_wer": 1.1740412979351031,
+      "step": 1600
+    }
+  ],
+  "max_steps": 1600,
+  "num_train_epochs": 320,
+  "total_flos": 6.632526100150112e+18,
+  "trial_name": null,
+  "trial_params": null
+}

xlsrDE-vlbg-L-attnDrp0.5/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eec94bb80bd146f474b68b2f356a2fb6a08cd5b391b2aa09e6c0dcf18a071438
+size 2799