End of training

README.md

@@ -3,6 +3,7 @@ library_name: transformers
 license: apache-2.0
 base_model: facebook/wav2vec2-large-xlsr-53
 tags:
+- automatic-speech-recognition
 - generated_from_trainer
 metrics:
 - wer
@@ -16,11 +17,11 @@ should probably proofread and complete it, then remove this comment. -->
 
 # wav2vec2-MYST-demo-dist
 
-This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on an unknown dataset.
+This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the MYST dataset.
 It achieves the following results on the evaluation set:
-- Loss: 22.9483
+- Loss: 19.7114
 - Wer: 1.0
-- Cer: 1.1343
+- Cer: 0.9951
 
 ## Model description
 

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 2.0,
+    "eval_cer": 0.99505515405097,
+    "eval_loss": 19.71142578125,
+    "eval_runtime": 1.7764,
+    "eval_samples": 55,
+    "eval_samples_per_second": 30.962,
+    "eval_steps_per_second": 2.252,
+    "eval_wer": 1.0,
+    "total_flos": 7.281788547445555e+16,
+    "train_loss": 16.014588832855225,
+    "train_runtime": 78.9646,
+    "train_samples": 192,
+    "train_samples_per_second": 4.863,
+    "train_steps_per_second": 0.608
+}

eval_results.json

@@ -0,0 +1,10 @@
+{
+    "epoch": 2.0,
+    "eval_cer": 0.99505515405097,
+    "eval_loss": 19.71142578125,
+    "eval_runtime": 1.7764,
+    "eval_samples": 55,
+    "eval_samples_per_second": 30.962,
+    "eval_steps_per_second": 2.252,
+    "eval_wer": 1.0
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 2.0,
+    "total_flos": 7.281788547445555e+16,
+    "train_loss": 16.014588832855225,
+    "train_runtime": 78.9646,
+    "train_samples": 192,
+    "train_samples_per_second": 4.863,
+    "train_steps_per_second": 0.608
+}

trainer_state.json

@@ -0,0 +1,388 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.0,
+  "eval_steps": 40,
+  "global_step": 48,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.041666666666666664,
+      "grad_norm": Infinity,
+      "learning_rate": 0.0,
+      "loss": 11.0247,
+      "step": 1
+    },
+    {
+      "epoch": 0.08333333333333333,
+      "grad_norm": Infinity,
+      "learning_rate": 0.0,
+      "loss": 12.7514,
+      "step": 2
+    },
+    {
+      "epoch": 0.125,
+      "grad_norm": 8.806501388549805,
+      "learning_rate": 6e-07,
+      "loss": 14.0529,
+      "step": 3
+    },
+    {
+      "epoch": 0.16666666666666666,
+      "grad_norm": Infinity,
+      "learning_rate": 6e-07,
+      "loss": 15.7234,
+      "step": 4
+    },
+    {
+      "epoch": 0.20833333333333334,
+      "grad_norm": Infinity,
+      "learning_rate": 6e-07,
+      "loss": 26.8783,
+      "step": 5
+    },
+    {
+      "epoch": 0.25,
+      "grad_norm": 20.22827911376953,
+      "learning_rate": 1.2e-06,
+      "loss": 24.4708,
+      "step": 6
+    },
+    {
+      "epoch": 0.2916666666666667,
+      "grad_norm": 7.415353775024414,
+      "learning_rate": 1.8e-06,
+      "loss": 11.0688,
+      "step": 7
+    },
+    {
+      "epoch": 0.3333333333333333,
+      "grad_norm": 7.202558517456055,
+      "learning_rate": 2.4e-06,
+      "loss": 11.3372,
+      "step": 8
+    },
+    {
+      "epoch": 0.375,
+      "grad_norm": 7.113807201385498,
+      "learning_rate": 2.9999999999999997e-06,
+      "loss": 11.8504,
+      "step": 9
+    },
+    {
+      "epoch": 0.4166666666666667,
+      "grad_norm": 7.7839155197143555,
+      "learning_rate": 3.6e-06,
+      "loss": 14.2548,
+      "step": 10
+    },
+    {
+      "epoch": 0.4583333333333333,
+      "grad_norm": 15.441122055053711,
+      "learning_rate": 4.2e-06,
+      "loss": 30.0114,
+      "step": 11
+    },
+    {
+      "epoch": 0.5,
+      "grad_norm": 10.832345008850098,
+      "learning_rate": 4.8e-06,
+      "loss": 18.9673,
+      "step": 12
+    },
+    {
+      "epoch": 0.5416666666666666,
+      "grad_norm": 8.285051345825195,
+      "learning_rate": 5.399999999999999e-06,
+      "loss": 11.0917,
+      "step": 13
+    },
+    {
+      "epoch": 0.5833333333333334,
+      "grad_norm": 8.244933128356934,
+      "learning_rate": 5.999999999999999e-06,
+      "loss": 12.1921,
+      "step": 14
+    },
+    {
+      "epoch": 0.625,
+      "grad_norm": 8.572517395019531,
+      "learning_rate": 6.599999999999999e-06,
+      "loss": 12.7475,
+      "step": 15
+    },
+    {
+      "epoch": 0.6666666666666666,
+      "grad_norm": 9.099493026733398,
+      "learning_rate": 7.2e-06,
+      "loss": 13.8911,
+      "step": 16
+    },
+    {
+      "epoch": 0.7083333333333334,
+      "grad_norm": 14.261921882629395,
+      "learning_rate": 7.799999999999998e-06,
+      "loss": 24.8105,
+      "step": 17
+    },
+    {
+      "epoch": 0.75,
+      "grad_norm": 7.782045841217041,
+      "learning_rate": 8.4e-06,
+      "loss": 15.1234,
+      "step": 18
+    },
+    {
+      "epoch": 0.7916666666666666,
+      "grad_norm": 9.227529525756836,
+      "learning_rate": 8.999999999999999e-06,
+      "loss": 13.6529,
+      "step": 19
+    },
+    {
+      "epoch": 0.8333333333333334,
+      "grad_norm": 7.739823818206787,
+      "learning_rate": 9.6e-06,
+      "loss": 11.3012,
+      "step": 20
+    },
+    {
+      "epoch": 0.875,
+      "grad_norm": 14.564947128295898,
+      "learning_rate": 1.02e-05,
+      "loss": 23.4074,
+      "step": 21
+    },
+    {
+      "epoch": 0.9166666666666666,
+      "grad_norm": 9.717215538024902,
+      "learning_rate": 1.0799999999999998e-05,
+      "loss": 16.9068,
+      "step": 22
+    },
+    {
+      "epoch": 0.9583333333333334,
+      "grad_norm": 9.182290077209473,
+      "learning_rate": 1.14e-05,
+      "loss": 17.5192,
+      "step": 23
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 15.464149475097656,
+      "learning_rate": 1.1999999999999999e-05,
+      "loss": 20.7789,
+      "step": 24
+    },
+    {
+      "epoch": 1.0416666666666667,
+      "grad_norm": 7.396650791168213,
+      "learning_rate": 1.26e-05,
+      "loss": 10.7157,
+      "step": 25
+    },
+    {
+      "epoch": 1.0833333333333333,
+      "grad_norm": 9.421314239501953,
+      "learning_rate": 1.3199999999999997e-05,
+      "loss": 14.0375,
+      "step": 26
+    },
+    {
+      "epoch": 1.125,
+      "grad_norm": 15.237442016601562,
+      "learning_rate": 1.3799999999999998e-05,
+      "loss": 24.223,
+      "step": 27
+    },
+    {
+      "epoch": 1.1666666666666667,
+      "grad_norm": 9.531441688537598,
+      "learning_rate": 1.44e-05,
+      "loss": 15.9401,
+      "step": 28
+    },
+    {
+      "epoch": 1.2083333333333333,
+      "grad_norm": 12.435009002685547,
+      "learning_rate": 1.4999999999999999e-05,
+      "loss": 23.032,
+      "step": 29
+    },
+    {
+      "epoch": 1.25,
+      "grad_norm": 14.035198211669922,
+      "learning_rate": 1.5599999999999996e-05,
+      "loss": 16.6483,
+      "step": 30
+    },
+    {
+      "epoch": 1.2916666666666667,
+      "grad_norm": 8.37588882446289,
+      "learning_rate": 1.6199999999999997e-05,
+      "loss": 12.122,
+      "step": 31
+    },
+    {
+      "epoch": 1.3333333333333333,
+      "grad_norm": 7.98301362991333,
+      "learning_rate": 1.68e-05,
+      "loss": 12.153,
+      "step": 32
+    },
+    {
+      "epoch": 1.375,
+      "grad_norm": 9.900574684143066,
+      "learning_rate": 1.74e-05,
+      "loss": 16.227,
+      "step": 33
+    },
+    {
+      "epoch": 1.4166666666666667,
+      "grad_norm": 6.607911109924316,
+      "learning_rate": 1.7999999999999997e-05,
+      "loss": 11.2944,
+      "step": 34
+    },
+    {
+      "epoch": 1.4583333333333333,
+      "grad_norm": 21.229183197021484,
+      "learning_rate": 1.8599999999999998e-05,
+      "loss": 27.8815,
+      "step": 35
+    },
+    {
+      "epoch": 1.5,
+      "grad_norm": 17.414100646972656,
+      "learning_rate": 1.92e-05,
+      "loss": 18.9613,
+      "step": 36
+    },
+    {
+      "epoch": 1.5416666666666665,
+      "grad_norm": 7.7176361083984375,
+      "learning_rate": 1.98e-05,
+      "loss": 11.2252,
+      "step": 37
+    },
+    {
+      "epoch": 1.5833333333333335,
+      "grad_norm": 6.465058326721191,
+      "learning_rate": 2.04e-05,
+      "loss": 9.7911,
+      "step": 38
+    },
+    {
+      "epoch": 1.625,
+      "grad_norm": 10.792200088500977,
+      "learning_rate": 2.1e-05,
+      "loss": 16.627,
+      "step": 39
+    },
+    {
+      "epoch": 1.6666666666666665,
+      "grad_norm": 5.8488054275512695,
+      "learning_rate": 2.1599999999999996e-05,
+      "loss": 10.3385,
+      "step": 40
+    },
+    {
+      "epoch": 1.6666666666666665,
+      "eval_cer": 1.1342715861544312,
+      "eval_loss": 22.94832420349121,
+      "eval_runtime": 1.8395,
+      "eval_samples_per_second": 29.899,
+      "eval_steps_per_second": 2.174,
+      "eval_wer": 1.0,
+      "step": 40
+    },
+    {
+      "epoch": 1.7083333333333335,
+      "grad_norm": 21.320125579833984,
+      "learning_rate": 2.2199999999999998e-05,
+      "loss": 24.4931,
+      "step": 41
+    },
+    {
+      "epoch": 1.75,
+      "grad_norm": 21.40454864501953,
+      "learning_rate": 2.28e-05,
+      "loss": 19.2591,
+      "step": 42
+    },
+    {
+      "epoch": 1.7916666666666665,
+      "grad_norm": 6.241701126098633,
+      "learning_rate": 2.34e-05,
+      "loss": 10.151,
+      "step": 43
+    },
+    {
+      "epoch": 1.8333333333333335,
+      "grad_norm": 10.232379913330078,
+      "learning_rate": 2.3999999999999997e-05,
+      "loss": 11.456,
+      "step": 44
+    },
+    {
+      "epoch": 1.875,
+      "grad_norm": 10.987034797668457,
+      "learning_rate": 2.4599999999999998e-05,
+      "loss": 10.9538,
+      "step": 45
+    },
+    {
+      "epoch": 1.9166666666666665,
+      "grad_norm": 24.182756423950195,
+      "learning_rate": 2.52e-05,
+      "loss": 16.68,
+      "step": 46
+    },
+    {
+      "epoch": 1.9583333333333335,
+      "grad_norm": 15.300764083862305,
+      "learning_rate": 2.5799999999999997e-05,
+      "loss": 13.9462,
+      "step": 47
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 19.339834213256836,
+      "learning_rate": 2.6399999999999995e-05,
+      "loss": 14.7293,
+      "step": 48
+    },
+    {
+      "epoch": 2.0,
+      "step": 48,
+      "total_flos": 7.281788547445555e+16,
+      "train_loss": 16.014588832855225,
+      "train_runtime": 78.9646,
+      "train_samples_per_second": 4.863,
+      "train_steps_per_second": 0.608
+    }
+  ],
+  "logging_steps": 1.0,
+  "max_steps": 48,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 400,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 7.281788547445555e+16,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}