Model save

README.md

@@ -1,6 +1,4 @@
 ---
-license: apache-2.0
-base_model: openai/whisper-tiny
 tags:
 - generated_from_trainer
 metrics:
@@ -15,10 +13,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # whisper-tiny-hi2_test
 
-This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the None dataset.
+This model was trained from scratch on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.5557
-- Wer: 60.5843
+- Loss: 0.4940
+- Wer: 59.7206
 
 ## Model description
 

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "openai/whisper-tiny",
+  "_name_or_path": "./whisper-tiny-hi2_test/whisper-tiny-hi-checkpoint-1",
   "activation_dropout": 0.0,
   "activation_function": "gelu",
   "apply_spec_augment": false,

runs/Jun26_08-55-05_bharatsahaiyak-test/events.out.tfevents.1719392105.bharatsahaiyak-test.1215638.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:165d0582673ef188b9d52b9c14d1a45c51671773f3298a7eb630d6fb6f82e50a
+size 6048

runs/Jun26_08-55-05_bharatsahaiyak-test/events.out.tfevents.1719392419.bharatsahaiyak-test.1215638.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:778b4630a2197cb19c0539dab2ab6c6ebbfe08a5a766bb35019cf7a58d314172
+size 406

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e3a2d63b5e8da641a253ed23f58db5d3da9941bebd2a70aef95105c81a4fa484
+oid sha256:d9287bb38beb16bf58664c3ed10f735c1bd05dd62bb178cf34a56cf97c52b277
 size 5240

whisper-tiny-hi-checkpoint-1/trainer_state.json

@@ -1,7 +1,7 @@
 {
-  "best_metric": 58.31498729889923,
-  "best_model_checkpoint": "./whisper-tiny-hi2_test/checkpoint-80",
-  "epoch": 6.25,
+  "best_metric": 59.720575783234544,
+  "best_model_checkpoint": "./whisper-tiny-hi2_test/checkpoint-120",
+  "epoch": 6.349206349206349,
   "eval_steps": 40,
   "global_step": 200,
   "is_hyper_param_search": false,
@@ -9,128 +9,128 @@
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.625,
-      "grad_norm": 8.544885635375977,
+      "epoch": 0.6349206349206349,
+      "grad_norm": 8.253315925598145,
       "learning_rate": 3.75e-05,
-      "loss": 1.8228,
+      "loss": 1.7555,
       "step": 20
     },
     {
-      "epoch": 1.25,
-      "grad_norm": 6.32118558883667,
+      "epoch": 1.2698412698412698,
+      "grad_norm": 6.022846698760986,
       "learning_rate": 3.75e-05,
-      "loss": 0.6803,
+      "loss": 0.6766,
       "step": 40
     },
     {
-      "epoch": 1.25,
-      "eval_loss": 0.6102083325386047,
-      "eval_runtime": 542.5987,
-      "eval_samples_per_second": 1.843,
-      "eval_steps_per_second": 1.843,
-      "eval_wer": 65.94411515664692,
+      "epoch": 1.2698412698412698,
+      "eval_loss": 0.6153799295425415,
+      "eval_runtime": 368.3001,
+      "eval_samples_per_second": 2.715,
+      "eval_steps_per_second": 0.679,
+      "eval_wer": 81.47332768839966,
       "step": 40
     },
     {
-      "epoch": 1.875,
-      "grad_norm": 6.4600510597229,
+      "epoch": 1.9047619047619047,
+      "grad_norm": 6.249065399169922,
       "learning_rate": 3.75e-05,
-      "loss": 0.4966,
+      "loss": 0.4907,
       "step": 60
     },
     {
-      "epoch": 2.5,
-      "grad_norm": 5.194849491119385,
+      "epoch": 2.5396825396825395,
+      "grad_norm": 5.881297588348389,
       "learning_rate": 3.75e-05,
-      "loss": 0.3649,
+      "loss": 0.3599,
       "step": 80
     },
     {
-      "epoch": 2.5,
-      "eval_loss": 0.5036975741386414,
-      "eval_runtime": 540.6408,
-      "eval_samples_per_second": 1.85,
-      "eval_steps_per_second": 1.85,
-      "eval_wer": 58.31498729889923,
+      "epoch": 2.5396825396825395,
+      "eval_loss": 0.5078147649765015,
+      "eval_runtime": 314.278,
+      "eval_samples_per_second": 3.182,
+      "eval_steps_per_second": 0.795,
+      "eval_wer": 67.01100762066045,
       "step": 80
     },
     {
-      "epoch": 3.125,
-      "grad_norm": 4.111001014709473,
+      "epoch": 3.1746031746031744,
+      "grad_norm": 4.2090606689453125,
       "learning_rate": 3.75e-05,
-      "loss": 0.3075,
+      "loss": 0.301,
       "step": 100
     },
     {
-      "epoch": 3.75,
-      "grad_norm": 4.358007431030273,
+      "epoch": 3.8095238095238093,
+      "grad_norm": 4.868574142456055,
       "learning_rate": 3.75e-05,
-      "loss": 0.2296,
+      "loss": 0.2297,
       "step": 120
     },
     {
-      "epoch": 3.75,
-      "eval_loss": 0.48966631293296814,
-      "eval_runtime": 538.04,
-      "eval_samples_per_second": 1.859,
-      "eval_steps_per_second": 1.859,
-      "eval_wer": 66.71464860287891,
+      "epoch": 3.8095238095238093,
+      "eval_loss": 0.49397116899490356,
+      "eval_runtime": 304.836,
+      "eval_samples_per_second": 3.28,
+      "eval_steps_per_second": 0.82,
+      "eval_wer": 59.720575783234544,
       "step": 120
     },
     {
-      "epoch": 4.375,
-      "grad_norm": 4.111566066741943,
+      "epoch": 4.444444444444445,
+      "grad_norm": 4.324865341186523,
       "learning_rate": 3.75e-05,
-      "loss": 0.1811,
+      "loss": 0.1754,
       "step": 140
     },
     {
-      "epoch": 5.0,
-      "grad_norm": 6.702174186706543,
+      "epoch": 5.079365079365079,
+      "grad_norm": 3.0074007511138916,
       "learning_rate": 3.75e-05,
-      "loss": 0.1564,
+      "loss": 0.153,
       "step": 160
     },
     {
-      "epoch": 5.0,
-      "eval_loss": 0.5130774974822998,
-      "eval_runtime": 529.7879,
-      "eval_samples_per_second": 1.888,
-      "eval_steps_per_second": 1.888,
-      "eval_wer": 65.34292972057578,
+      "epoch": 5.079365079365079,
+      "eval_loss": 0.5192885994911194,
+      "eval_runtime": 309.7886,
+      "eval_samples_per_second": 3.228,
+      "eval_steps_per_second": 0.807,
+      "eval_wer": 62.07451312447078,
       "step": 160
     },
     {
-      "epoch": 5.625,
-      "grad_norm": 3.0882856845855713,
+      "epoch": 5.714285714285714,
+      "grad_norm": 2.9954638481140137,
       "learning_rate": 3.75e-05,
-      "loss": 0.101,
+      "loss": 0.1049,
       "step": 180
     },
     {
-      "epoch": 6.25,
-      "grad_norm": 4.524318695068359,
+      "epoch": 6.349206349206349,
+      "grad_norm": 3.4090383052825928,
       "learning_rate": 3.75e-05,
-      "loss": 0.0905,
+      "loss": 0.0885,
       "step": 200
     },
     {
-      "epoch": 6.25,
-      "eval_loss": 0.5503659248352051,
-      "eval_runtime": 587.5545,
-      "eval_samples_per_second": 1.702,
-      "eval_steps_per_second": 1.702,
-      "eval_wer": 63.903471634208294,
+      "epoch": 6.349206349206349,
+      "eval_loss": 0.5557394027709961,
+      "eval_runtime": 308.2851,
+      "eval_samples_per_second": 3.244,
+      "eval_steps_per_second": 0.811,
+      "eval_wer": 60.584250635055035,
       "step": 200
     },
     {
-      "epoch": 6.25,
+      "epoch": 6.349206349206349,
       "step": 200,
-      "total_flos": 1.5401574531072e+17,
-      "train_loss": 0.44307032763957976,
-      "train_runtime": 3561.8571,
-      "train_samples_per_second": 1.797,
-      "train_steps_per_second": 0.056
+      "total_flos": 1.5637915828224e+17,
+      "train_loss": 0.4335097998380661,
+      "train_runtime": 2402.3095,
+      "train_samples_per_second": 2.664,
+      "train_steps_per_second": 0.083
     }
   ],
   "logging_steps": 20,
@@ -150,8 +150,8 @@
       "attributes": {}
     }
   },
-  "total_flos": 1.5401574531072e+17,
-  "train_batch_size": 32,
+  "total_flos": 1.5637915828224e+17,
+  "train_batch_size": 16,
   "trial_name": null,
   "trial_params": null
 }

whisper-tiny-hi-checkpoint-2/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:802a25ee7db91faba0d8c4d8ea560053183efcadf455712cb855292076f680e4
+size 151099494

whisper-tiny-hi-checkpoint-2/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:549ce3613e0053502a77050b29a76987037c7c15586538493db5d847eb76fe1c
+oid sha256:d9287bb38beb16bf58664c3ed10f735c1bd05dd62bb178cf34a56cf97c52b277
 size 5240