End of training

README.md

@@ -1,15 +1,20 @@
 ---
 library_name: transformers
+language:
+- jv
 license: apache-2.0
 base_model: openai/whisper-tiny
 tags:
+- whisper
+- javanese
+- asr
 - generated_from_trainer
 datasets:
 - jv_id_asr_split
 metrics:
 - wer
 model-index:
-- name: whisper-tiny-javanese-openslr-v2
+- name: Whisper Tiny Java
   results:
   - task:
       name: Automatic Speech Recognition
@@ -23,18 +28,18 @@ model-index:
     metrics:
     - name: Wer
       type: wer
-      value: 0.6603038810125197
+      value: 0.4824482518630893
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 should probably proofread and complete it, then remove this comment. -->
 
-# whisper-tiny-javanese-openslr-v2
+# Whisper Tiny Java
 
 This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the jv_id_asr_split dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.3406
-- Wer: 0.6603
+- Loss: 0.5638
+- Wer: 0.4824
 
 ## Model description
 

all_results.json

@@ -1,10 +1,10 @@
 {
     "epoch": 4.9922212618841835,
     "eval_loss": 0.5638014078140259,
-    "eval_runtime": 4176.5631,
+    "eval_runtime": 4431.4054,
     "eval_samples": 18504,
-    "eval_samples_per_second": 4.43,
-    "eval_steps_per_second": 0.277,
+    "eval_samples_per_second": 4.176,
+    "eval_steps_per_second": 0.261,
     "eval_wer": 0.4824482518630893,
     "pretrained_eval_loss": 5.0976738929748535,
     "pretrained_eval_model_preparation_time": 0.0025,
@@ -13,9 +13,9 @@
     "pretrained_eval_steps_per_second": 0.21,
     "pretrained_eval_wer": 1.3043423861619188,
     "total_flos": 1.819709817421824e+19,
-    "train_loss": 0.5687196460974671,
-    "train_runtime": 175635.8061,
+    "train_loss": 0.0,
+    "train_runtime": 0.166,
     "train_samples": 148052,
-    "train_samples_per_second": 4.215,
-    "train_steps_per_second": 0.016
+    "train_samples_per_second": 4460643.449,
+    "train_steps_per_second": 17414.502
 }

eval_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 4.9922212618841835,
     "eval_loss": 0.5638014078140259,
-    "eval_runtime": 4176.5631,
+    "eval_runtime": 4431.4054,
     "eval_samples": 18504,
-    "eval_samples_per_second": 4.43,
-    "eval_steps_per_second": 0.277,
+    "eval_samples_per_second": 4.176,
+    "eval_steps_per_second": 0.261,
     "eval_wer": 0.4824482518630893
 }

runs/Mar12_00-38-44_dgx-a100/events.out.tfevents.1741719184.dgx-a100.2122149.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:45550cfd870e9e9e7c0760638b5a84c984359e613496709e315726c4dc2dcec4
+size 406

train_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 4.9922212618841835,
     "total_flos": 1.819709817421824e+19,
-    "train_loss": 0.5687196460974671,
-    "train_runtime": 175635.8061,
+    "train_loss": 0.0,
+    "train_runtime": 0.166,
     "train_samples": 148052,
-    "train_samples_per_second": 4.215,
-    "train_steps_per_second": 0.016
+    "train_samples_per_second": 4460643.449,
+    "train_steps_per_second": 17414.502
 }

trainer_state.json

@@ -253,10 +253,10 @@
       "epoch": 4.9922212618841835,
       "step": 2890,
       "total_flos": 1.819709817421824e+19,
-      "train_loss": 0.5687196460974671,
-      "train_runtime": 175635.8061,
-      "train_samples_per_second": 4.215,
-      "train_steps_per_second": 0.016
+      "train_loss": 0.0,
+      "train_runtime": 0.166,
+      "train_samples_per_second": 4460643.449,
+      "train_steps_per_second": 17414.502
     }
   ],
   "logging_steps": 100,