End of training

README.md

@@ -3,6 +3,8 @@ library_name: transformers
 license: apache-2.0
 base_model: facebook/wav2vec2-xls-r-1b
 tags:
+- automatic-speech-recognition
+- natbed
 - generated_from_trainer
 metrics:
 - wer
@@ -16,10 +18,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # xls-r-1b-bem-natbed-combined-model
 
-This model is a fine-tuned version of [facebook/wav2vec2-xls-r-1b](https://huggingface.co/facebook/wav2vec2-xls-r-1b) on an unknown dataset.
+This model is a fine-tuned version of [facebook/wav2vec2-xls-r-1b](https://huggingface.co/facebook/wav2vec2-xls-r-1b) on the NATBED - BEM dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.7916
-- Wer: 0.7758
+- Loss: 0.7801
+- Wer: 0.7879
 
 ## Model description
 

all_results.json

@@ -0,0 +1,15 @@
+{
+    "epoch": 2.3794614902943017,
+    "eval_loss": 0.780085027217865,
+    "eval_runtime": 90.2213,
+    "eval_samples": 1358,
+    "eval_samples_per_second": 15.052,
+    "eval_steps_per_second": 1.884,
+    "eval_wer": 0.7878639188747137,
+    "total_flos": 9.705775703537089e+18,
+    "train_loss": 1.14412841796875,
+    "train_runtime": 3792.6045,
+    "train_samples": 6387,
+    "train_samples_per_second": 50.522,
+    "train_steps_per_second": 6.312
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 2.3794614902943017,
+    "eval_loss": 0.780085027217865,
+    "eval_runtime": 90.2213,
+    "eval_samples": 1358,
+    "eval_samples_per_second": 15.052,
+    "eval_steps_per_second": 1.884,
+    "eval_wer": 0.7878639188747137
+}

runs/Sep29_08-03-03_08560ebc7a23/events.out.tfevents.1727600998.08560ebc7a23.5836.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f75875bcc0eed12510ed705b33469727371d453f994fbb6360e274f520a56420
+size 406

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 2.3794614902943017,
+    "total_flos": 9.705775703537089e+18,
+    "train_loss": 1.14412841796875,
+    "train_runtime": 3792.6045,
+    "train_samples": 6387,
+    "train_samples_per_second": 50.522,
+    "train_steps_per_second": 6.312
+}

trainer_state.json

@@ -0,0 +1,243 @@
+{
+  "best_metric": 0.780085027217865,
+  "best_model_checkpoint": "./xls-r-1b-bem-natbed-combined-model/checkpoint-1600",
+  "epoch": 2.3794614902943017,
+  "eval_steps": 100,
+  "global_step": 1900,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.12523481527864747,
+      "eval_loss": 1.372534990310669,
+      "eval_runtime": 91.0174,
+      "eval_samples_per_second": 14.92,
+      "eval_steps_per_second": 1.868,
+      "eval_wer": 0.9804546941445862,
+      "step": 100
+    },
+    {
+      "epoch": 0.25046963055729493,
+      "eval_loss": 0.9491279721260071,
+      "eval_runtime": 89.1948,
+      "eval_samples_per_second": 15.225,
+      "eval_steps_per_second": 1.906,
+      "eval_wer": 0.8541053320248609,
+      "step": 200
+    },
+    {
+      "epoch": 0.37570444583594237,
+      "eval_loss": 0.9524340033531189,
+      "eval_runtime": 89.2642,
+      "eval_samples_per_second": 15.213,
+      "eval_steps_per_second": 1.904,
+      "eval_wer": 0.8137062479555119,
+      "step": 300
+    },
+    {
+      "epoch": 0.5009392611145899,
+      "eval_loss": 1.0354979038238525,
+      "eval_runtime": 88.9821,
+      "eval_samples_per_second": 15.262,
+      "eval_steps_per_second": 1.91,
+      "eval_wer": 0.9016192345436702,
+      "step": 400
+    },
+    {
+      "epoch": 0.6261740763932373,
+      "grad_norm": 3.467607259750366,
+      "learning_rate": 0.00029759999999999997,
+      "loss": 1.7945,
+      "step": 500
+    },
+    {
+      "epoch": 0.6261740763932373,
+      "eval_loss": 0.9610921144485474,
+      "eval_runtime": 89.3241,
+      "eval_samples_per_second": 15.203,
+      "eval_steps_per_second": 1.903,
+      "eval_wer": 0.8788027477919529,
+      "step": 500
+    },
+    {
+      "epoch": 0.7514088916718847,
+      "eval_loss": 0.9790019392967224,
+      "eval_runtime": 88.7534,
+      "eval_samples_per_second": 15.301,
+      "eval_steps_per_second": 1.915,
+      "eval_wer": 0.864245992803402,
+      "step": 600
+    },
+    {
+      "epoch": 0.8766437069505323,
+      "eval_loss": 0.9877487421035767,
+      "eval_runtime": 88.7715,
+      "eval_samples_per_second": 15.298,
+      "eval_steps_per_second": 1.915,
+      "eval_wer": 0.8602387962054302,
+      "step": 700
+    },
+    {
+      "epoch": 1.0018785222291797,
+      "eval_loss": 0.9603913426399231,
+      "eval_runtime": 88.9263,
+      "eval_samples_per_second": 15.271,
+      "eval_steps_per_second": 1.912,
+      "eval_wer": 0.8925417075564279,
+      "step": 800
+    },
+    {
+      "epoch": 1.127113337507827,
+      "eval_loss": 0.8880479335784912,
+      "eval_runtime": 90.5701,
+      "eval_samples_per_second": 14.994,
+      "eval_steps_per_second": 1.877,
+      "eval_wer": 0.8327608766764802,
+      "step": 900
+    },
+    {
+      "epoch": 1.2523481527864746,
+      "grad_norm": 1.2925059795379639,
+      "learning_rate": 0.00029365187713310576,
+      "loss": 0.9885,
+      "step": 1000
+    },
+    {
+      "epoch": 1.2523481527864746,
+      "eval_loss": 0.8917101621627808,
+      "eval_runtime": 89.0247,
+      "eval_samples_per_second": 15.254,
+      "eval_steps_per_second": 1.91,
+      "eval_wer": 0.8368498527968596,
+      "step": 1000
+    },
+    {
+      "epoch": 1.3775829680651221,
+      "eval_loss": 0.9034265875816345,
+      "eval_runtime": 89.348,
+      "eval_samples_per_second": 15.199,
+      "eval_steps_per_second": 1.903,
+      "eval_wer": 0.8306346090938829,
+      "step": 1100
+    },
+    {
+      "epoch": 1.5028177833437697,
+      "eval_loss": 0.8478356003761292,
+      "eval_runtime": 89.5799,
+      "eval_samples_per_second": 15.16,
+      "eval_steps_per_second": 1.898,
+      "eval_wer": 0.7937520444880601,
+      "step": 1200
+    },
+    {
+      "epoch": 1.628052598622417,
+      "eval_loss": 0.8666252493858337,
+      "eval_runtime": 89.9747,
+      "eval_samples_per_second": 15.093,
+      "eval_steps_per_second": 1.889,
+      "eval_wer": 0.8627739614000655,
+      "step": 1300
+    },
+    {
+      "epoch": 1.7532874139010644,
+      "eval_loss": 0.833095371723175,
+      "eval_runtime": 89.2147,
+      "eval_samples_per_second": 15.222,
+      "eval_steps_per_second": 1.906,
+      "eval_wer": 0.8218024206738632,
+      "step": 1400
+    },
+    {
+      "epoch": 1.878522229179712,
+      "grad_norm": 7.0910844802856445,
+      "learning_rate": 0.0002872525597269624,
+      "loss": 0.8854,
+      "step": 1500
+    },
+    {
+      "epoch": 1.878522229179712,
+      "eval_loss": 0.840486466884613,
+      "eval_runtime": 89.4023,
+      "eval_samples_per_second": 15.19,
+      "eval_steps_per_second": 1.902,
+      "eval_wer": 0.8045469414458619,
+      "step": 1500
+    },
+    {
+      "epoch": 2.0037570444583594,
+      "eval_loss": 0.780085027217865,
+      "eval_runtime": 91.0684,
+      "eval_samples_per_second": 14.912,
+      "eval_steps_per_second": 1.867,
+      "eval_wer": 0.7878639188747137,
+      "step": 1600
+    },
+    {
+      "epoch": 2.128991859737007,
+      "eval_loss": 0.8304778933525085,
+      "eval_runtime": 89.8902,
+      "eval_samples_per_second": 15.107,
+      "eval_steps_per_second": 1.891,
+      "eval_wer": 0.7917075564278705,
+      "step": 1700
+    },
+    {
+      "epoch": 2.254226675015654,
+      "eval_loss": 0.7971779108047485,
+      "eval_runtime": 89.8082,
+      "eval_samples_per_second": 15.121,
+      "eval_steps_per_second": 1.893,
+      "eval_wer": 0.7910533202486097,
+      "step": 1800
+    },
+    {
+      "epoch": 2.3794614902943017,
+      "eval_loss": 0.7915882468223572,
+      "eval_runtime": 90.3859,
+      "eval_samples_per_second": 15.024,
+      "eval_steps_per_second": 1.881,
+      "eval_wer": 0.7758423290807982,
+      "step": 1900
+    },
+    {
+      "epoch": 2.3794614902943017,
+      "step": 1900,
+      "total_flos": 9.705775703537089e+18,
+      "train_loss": 1.14412841796875,
+      "train_runtime": 3792.6045,
+      "train_samples_per_second": 50.522,
+      "train_steps_per_second": 6.312
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 23940,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 30,
+  "save_steps": 400,
+  "stateful_callbacks": {
+    "EarlyStoppingCallback": {
+      "args": {
+        "early_stopping_patience": 3,
+        "early_stopping_threshold": 0.0
+      },
+      "attributes": {
+        "early_stopping_patience_counter": 0
+      }
+    },
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 9.705775703537089e+18,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}