Instructions to use alakxender/whisper-large-v3-cv17-dv with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use alakxender/whisper-large-v3-cv17-dv with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="alakxender/whisper-large-v3-cv17-dv")# Load model directly from transformers import AutoProcessor, AutoModelForSpeechSeq2Seq processor = AutoProcessor.from_pretrained("alakxender/whisper-large-v3-cv17-dv") model = AutoModelForSpeechSeq2Seq.from_pretrained("alakxender/whisper-large-v3-cv17-dv") - Notebooks
- Google Colab
- Kaggle
Commit ·
5fe9b84
1
Parent(s): 9479036
Training in progress, step 3300, checkpoint
Browse files
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": 71.37931034482759,
|
| 3 |
"best_model_checkpoint": "./whisper-large-v3-cv17-dv/checkpoint-2700",
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 300,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -307,6 +307,36 @@
|
|
| 307 |
"eval_steps_per_second": 0.021,
|
| 308 |
"eval_wer": 71.72413793103448,
|
| 309 |
"step": 3000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 310 |
}
|
| 311 |
],
|
| 312 |
"logging_steps": 100,
|
|
@@ -326,7 +356,7 @@
|
|
| 326 |
"attributes": {}
|
| 327 |
}
|
| 328 |
},
|
| 329 |
-
"total_flos": 1.
|
| 330 |
"train_batch_size": 16,
|
| 331 |
"trial_name": null,
|
| 332 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": 71.37931034482759,
|
| 3 |
"best_model_checkpoint": "./whisper-large-v3-cv17-dv/checkpoint-2700",
|
| 4 |
+
"epoch": 9.908794788273616,
|
| 5 |
"eval_steps": 300,
|
| 6 |
+
"global_step": 3300,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 307 |
"eval_steps_per_second": 0.021,
|
| 308 |
"eval_wer": 71.72413793103448,
|
| 309 |
"step": 3000
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 9.257328990228013,
|
| 313 |
+
"grad_norm": 0.23316721618175507,
|
| 314 |
+
"learning_rate": 2.5857142857142856e-06,
|
| 315 |
+
"loss": 0.0031,
|
| 316 |
+
"step": 3100
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 9.583061889250814,
|
| 320 |
+
"grad_norm": 0.7372754216194153,
|
| 321 |
+
"learning_rate": 2.3000000000000004e-06,
|
| 322 |
+
"loss": 0.0027,
|
| 323 |
+
"step": 3200
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 9.908794788273616,
|
| 327 |
+
"grad_norm": 0.2684876322746277,
|
| 328 |
+
"learning_rate": 2.0142857142857144e-06,
|
| 329 |
+
"loss": 0.0023,
|
| 330 |
+
"step": 3300
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 9.908794788273616,
|
| 334 |
+
"eval_loss": 0.42460787296295166,
|
| 335 |
+
"eval_runtime": 146.943,
|
| 336 |
+
"eval_samples_per_second": 0.681,
|
| 337 |
+
"eval_steps_per_second": 0.02,
|
| 338 |
+
"eval_wer": 72.58620689655172,
|
| 339 |
+
"step": 3300
|
| 340 |
}
|
| 341 |
],
|
| 342 |
"logging_steps": 100,
|
|
|
|
| 356 |
"attributes": {}
|
| 357 |
}
|
| 358 |
},
|
| 359 |
+
"total_flos": 1.791840606879744e+20,
|
| 360 |
"train_batch_size": 16,
|
| 361 |
"trial_name": null,
|
| 362 |
"trial_params": null
|