End of training

README.md

@@ -3,11 +3,23 @@ license: apache-2.0
 base_model: openai/whisper-tiny
 tags:
 - generated_from_trainer
+datasets:
+- KasuleTrevor/lingala_20hr
 metrics:
 - wer
 model-index:
 - name: whisper-ling-asr
-  results: []
+  results:
+  - task:
+      name: Automatic Speech Recognition
+      type: automatic-speech-recognition
+    dataset:
+      name: KasuleTrevor/lingala_20hr
+      type: KasuleTrevor/lingala_20hr
+    metrics:
+    - name: Wer
+      type: wer
+      value: 0.2979101583539804
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -16,10 +28,10 @@ should probably proofread and complete it, then remove this comment. -->
 [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/abdouaziz/wolof-asr/runs/4gbjp7dq)
 # whisper-ling-asr
 
-This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on an unknown dataset.
+This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the KasuleTrevor/lingala_20hr dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.5673
-- Wer: 0.2177
+- Loss: 0.4835
+- Wer: 0.2979
 
 ## Model description
 

all_results.json

@@ -0,0 +1,15 @@
+{
+    "epoch": 14.925373134328359,
+    "eval_loss": 0.4834844172000885,
+    "eval_runtime": 473.9304,
+    "eval_samples": 1099,
+    "eval_samples_per_second": 2.319,
+    "eval_steps_per_second": 0.146,
+    "eval_wer": 0.2979101583539804,
+    "total_flos": 1.96882148081664e+18,
+    "train_loss": 0.1769935465335846,
+    "train_runtime": 17207.4544,
+    "train_samples": 5358,
+    "train_samples_per_second": 11.158,
+    "train_steps_per_second": 0.697
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 14.925373134328359,
+    "eval_loss": 0.4834844172000885,
+    "eval_runtime": 473.9304,
+    "eval_samples": 1099,
+    "eval_samples_per_second": 2.319,
+    "eval_steps_per_second": 0.146,
+    "eval_wer": 0.2979101583539804
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 14.925373134328359,
+    "total_flos": 1.96882148081664e+18,
+    "train_loss": 0.1769935465335846,
+    "train_runtime": 17207.4544,
+    "train_samples": 5358,
+    "train_samples_per_second": 11.158,
+    "train_steps_per_second": 0.697
+}

trainer_state.json

@@ -0,0 +1,166 @@
+{
+  "best_metric": 0.4834844172000885,
+  "best_model_checkpoint": "whisper-ling-asr/checkpoint-2000",
+  "epoch": 14.925373134328359,
+  "eval_steps": 1000,
+  "global_step": 5000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.4925373134328357,
+      "grad_norm": 7.594086647033691,
+      "learning_rate": 9.625941422594142e-06,
+      "loss": 0.8297,
+      "step": 500
+    },
+    {
+      "epoch": 2.9850746268656714,
+      "grad_norm": 8.005925178527832,
+      "learning_rate": 9.207531380753139e-06,
+      "loss": 0.3703,
+      "step": 1000
+    },
+    {
+      "epoch": 2.9850746268656714,
+      "eval_loss": 0.5100315809249878,
+      "eval_runtime": 410.9956,
+      "eval_samples_per_second": 2.674,
+      "eval_steps_per_second": 0.168,
+      "eval_wer": 0.27345685661962726,
+      "step": 1000
+    },
+    {
+      "epoch": 4.477611940298507,
+      "grad_norm": 4.732856273651123,
+      "learning_rate": 8.789121338912134e-06,
+      "loss": 0.2182,
+      "step": 1500
+    },
+    {
+      "epoch": 5.970149253731344,
+      "grad_norm": 4.545703411102295,
+      "learning_rate": 8.37071129707113e-06,
+      "loss": 0.1487,
+      "step": 2000
+    },
+    {
+      "epoch": 5.970149253731344,
+      "eval_loss": 0.4834844172000885,
+      "eval_runtime": 462.8067,
+      "eval_samples_per_second": 2.375,
+      "eval_steps_per_second": 0.149,
+      "eval_wer": 0.2979101583539804,
+      "step": 2000
+    },
+    {
+      "epoch": 7.462686567164179,
+      "grad_norm": 3.4771857261657715,
+      "learning_rate": 7.953138075313808e-06,
+      "loss": 0.0874,
+      "step": 2500
+    },
+    {
+      "epoch": 8.955223880597014,
+      "grad_norm": 2.7059333324432373,
+      "learning_rate": 7.534728033472803e-06,
+      "loss": 0.0556,
+      "step": 3000
+    },
+    {
+      "epoch": 8.955223880597014,
+      "eval_loss": 0.5049698948860168,
+      "eval_runtime": 404.4404,
+      "eval_samples_per_second": 2.717,
+      "eval_steps_per_second": 0.171,
+      "eval_wer": 0.254389744694603,
+      "step": 3000
+    },
+    {
+      "epoch": 10.447761194029852,
+      "grad_norm": 2.1922836303710938,
+      "learning_rate": 7.116317991631799e-06,
+      "loss": 0.0295,
+      "step": 3500
+    },
+    {
+      "epoch": 11.940298507462687,
+      "grad_norm": 0.999194324016571,
+      "learning_rate": 6.697907949790795e-06,
+      "loss": 0.0171,
+      "step": 4000
+    },
+    {
+      "epoch": 11.940298507462687,
+      "eval_loss": 0.5400618314743042,
+      "eval_runtime": 374.2093,
+      "eval_samples_per_second": 2.937,
+      "eval_steps_per_second": 0.184,
+      "eval_wer": 0.23704621350856403,
+      "step": 4000
+    },
+    {
+      "epoch": 13.432835820895523,
+      "grad_norm": 1.5301833152770996,
+      "learning_rate": 6.279497907949791e-06,
+      "loss": 0.0084,
+      "step": 4500
+    },
+    {
+      "epoch": 14.925373134328359,
+      "grad_norm": 0.46232014894485474,
+      "learning_rate": 5.8610878661087865e-06,
+      "loss": 0.0051,
+      "step": 5000
+    },
+    {
+      "epoch": 14.925373134328359,
+      "eval_loss": 0.5673447251319885,
+      "eval_runtime": 355.0078,
+      "eval_samples_per_second": 3.096,
+      "eval_steps_per_second": 0.194,
+      "eval_wer": 0.21765593019498006,
+      "step": 5000
+    },
+    {
+      "epoch": 14.925373134328359,
+      "step": 5000,
+      "total_flos": 1.96882148081664e+18,
+      "train_loss": 0.1769935465335846,
+      "train_runtime": 17207.4544,
+      "train_samples_per_second": 11.158,
+      "train_steps_per_second": 0.697
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 12000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 36,
+  "save_steps": 1000,
+  "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": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.96882148081664e+18,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}