End of training

README.md

@@ -19,7 +19,7 @@ model-index:
     metrics:
     - name: Wer
       type: wer
-      value: 24.30636512007461
+      value: 24.014921893215202
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -29,8 +29,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the lyhourt/clean_6 dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.2900
-- Wer: 24.3064
+- Loss: 0.2886
+- Wer: 24.0149
 
 ## Model description
 
@@ -56,7 +56,7 @@ The following hyperparameters were used during training:
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_steps: 50
-- training_steps: 800
+- training_steps: 1000
 - mixed_precision_training: Native AMP
 
 ### Training results
@@ -67,6 +67,7 @@ The following hyperparameters were used during training:
 | 0.1827        | 1.0    | 400  | 0.2953          | 24.7144 |
 | 0.0907        | 1.1342 | 600  | 0.2921          | 24.3413 |
 | 0.0904        | 1.5123 | 800  | 0.2900          | 24.3064 |
+| 0.0823        | 1.8904 | 1000 | 0.2886          | 24.0149 |
 
 
 ### Framework versions

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:da4e39516fd2c574cde049cc323a39dd5de082636bac569f33e739a6369e29c3
+oid sha256:e72923ad30f0957f11a615e18eefb023aa94497336fc831a2f8b9fa384853875
 size 966995080

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dbd06e8492cf5f9098994eae7379bdec4b8a992134de50c159b29f2af7d36e16
+oid sha256:309d78c0cbac356af547d46b1e70776079ec1d34f37e987fe5340ad766fae6b9
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:345bea270db4afa20c0b04262d1b80fec7c4873e001e3e4c19066324a81d9415
+oid sha256:888e6a44b40497069d2e43e76c43808360f08b18da088d54334ffdb48c5b1b0a
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 24.014921893215202,
-  "best_model_checkpoint": "./whisper-small-clean_6-v4/checkpoint-1000",
-  "epoch": 1.8903591682419658,
+  "best_metric": 24.30636512007461,
+  "best_model_checkpoint": "./whisper-small-clean_6-v4/checkpoint-800",
+  "epoch": 1.5122873345935728,
   "eval_steps": 200,
-  "global_step": 1000,
+  "global_step": 800,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -267,75 +267,10 @@
       "eval_steps_per_second": 0.076,
       "eval_wer": 24.30636512007461,
       "step": 800
-    },
-    {
-      "epoch": 1.559546313799622,
-      "grad_norm": 2.5037174224853516,
-      "learning_rate": 1.8421052631578948e-06,
-      "loss": 0.0912,
-      "step": 825
-    },
-    {
-      "epoch": 1.606805293005671,
-      "grad_norm": 3.10439395904541,
-      "learning_rate": 1.5789473684210526e-06,
-      "loss": 0.0876,
-      "step": 850
-    },
-    {
-      "epoch": 1.6540642722117203,
-      "grad_norm": 2.7944157123565674,
-      "learning_rate": 1.3157894736842106e-06,
-      "loss": 0.0877,
-      "step": 875
-    },
-    {
-      "epoch": 1.7013232514177694,
-      "grad_norm": 2.672607421875,
-      "learning_rate": 1.0526315789473685e-06,
-      "loss": 0.0934,
-      "step": 900
-    },
-    {
-      "epoch": 1.7485822306238186,
-      "grad_norm": 2.3285105228424072,
-      "learning_rate": 7.894736842105263e-07,
-      "loss": 0.0888,
-      "step": 925
-    },
-    {
-      "epoch": 1.7958412098298677,
-      "grad_norm": 2.0739948749542236,
-      "learning_rate": 5.263157894736843e-07,
-      "loss": 0.0974,
-      "step": 950
-    },
-    {
-      "epoch": 1.8431001890359169,
-      "grad_norm": 2.6488840579986572,
-      "learning_rate": 2.6315789473684213e-07,
-      "loss": 0.0874,
-      "step": 975
-    },
-    {
-      "epoch": 1.8903591682419658,
-      "grad_norm": 2.463960886001587,
-      "learning_rate": 0.0,
-      "loss": 0.0823,
-      "step": 1000
-    },
-    {
-      "epoch": 1.8903591682419658,
-      "eval_loss": 0.2885694205760956,
-      "eval_runtime": 632.9247,
-      "eval_samples_per_second": 2.409,
-      "eval_steps_per_second": 0.076,
-      "eval_wer": 24.014921893215202,
-      "step": 1000
     }
   ],
   "logging_steps": 25,
-  "max_steps": 1000,
+  "max_steps": 800,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 2,
   "save_steps": 200,
@@ -351,7 +286,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1.850784747872256e+19,
+  "total_flos": 1.481395435462656e+19,
   "train_batch_size": 64,
   "trial_name": null,
   "trial_params": null

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ae62acf766aec21814d7588bec18284ba3498b083aa6cdd95022c9bf4a540e50
+oid sha256:6d7915e08ab8f42cb0e587a2e8d8dc8bb8879d7d74ee37b8524edf8f0faf9041
 size 5240