End of training

README.md

@@ -1,39 +1,42 @@
 ---
+language:
+- id
 license: apache-2.0
 base_model: openai/whisper-tiny
 tags:
+- whisper-event
 - generated_from_trainer
 datasets:
-- common_voice_16_0
+- mozilla-foundation/common_voice_16_0
 metrics:
 - wer
 model-index:
-- name: breeze-dsw-tiny-id
+- name: Breeze DSW Indonesian - tiny
   results:
   - task:
       name: Automatic Speech Recognition
       type: automatic-speech-recognition
     dataset:
-      name: common_voice_16_0
-      type: common_voice_16_0
+      name: mozilla-foundation/common_voice_16_0 id
+      type: mozilla-foundation/common_voice_16_0
       config: id
       split: test
       args: id
     metrics:
     - name: Wer
       type: wer
-      value: 43.95528567485509
+      value: 43.44465912227436
 ---
 
 <!-- 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. -->
 
-# breeze-dsw-tiny-id
+# Breeze DSW Indonesian - tiny
 
-This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the common_voice_16_0 dataset.
+This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the mozilla-foundation/common_voice_16_0 id dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.7134
-- Wer: 43.9553
+- Loss: 0.7090
+- Wer: 43.4447
 
 ## Model description
 

all_results.json

@@ -1,12 +1,12 @@
 {
-    "epoch": 3.03,
-    "eval_loss": 0.7109375,
-    "eval_runtime": 1127.2337,
-    "eval_samples_per_second": 3.231,
-    "eval_steps_per_second": 0.202,
-    "eval_wer": 45.243352654338025,
-    "train_loss": 0.0,
-    "train_runtime": 67.5132,
-    "train_samples_per_second": 947.963,
-    "train_steps_per_second": 29.624
+    "epoch": 6.06,
+    "eval_loss": 0.708984375,
+    "eval_runtime": 1112.1656,
+    "eval_samples_per_second": 3.275,
+    "eval_steps_per_second": 0.205,
+    "eval_wer": 43.44465912227436,
+    "train_loss": 0.5104879150390625,
+    "train_runtime": 24718.0782,
+    "train_samples_per_second": 1.295,
+    "train_steps_per_second": 0.04
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 3.03,
-    "eval_loss": 0.7109375,
-    "eval_runtime": 1127.2337,
-    "eval_samples_per_second": 3.231,
-    "eval_steps_per_second": 0.202,
-    "eval_wer": 45.243352654338025
+    "epoch": 6.06,
+    "eval_loss": 0.708984375,
+    "eval_runtime": 1112.1656,
+    "eval_samples_per_second": 3.275,
+    "eval_steps_per_second": 0.205,
+    "eval_wer": 43.44465912227436
 }

runs/Jan13_12-02-57_knight/events.out.tfevents.1705169664.knight.21364.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:625605ab44cd33c469bfd3a51d8b4ff10635f8a0c8409b9a6caccadb4fc31b4e
+size 406

train_results.json

@@ -1,7 +1,7 @@
 {
-    "epoch": 3.03,
-    "train_loss": 0.0,
-    "train_runtime": 67.5132,
-    "train_samples_per_second": 947.963,
-    "train_steps_per_second": 29.624
+    "epoch": 6.06,
+    "train_loss": 0.5104879150390625,
+    "train_runtime": 24718.0782,
+    "train_samples_per_second": 1.295,
+    "train_steps_per_second": 0.04
 }

trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 45.243352654338025,
-  "best_model_checkpoint": "/cosmos/home/sp-operator/ai/training/models/huggingface/scripts/../breeze-dsw-tiny-id/checkpoint-500",
-  "epoch": 3.032,
+  "best_metric": 43.44465912227436,
+  "best_model_checkpoint": "/cosmos/home/sp-operator/ai/training/models/huggingface/scripts/../breeze-dsw-tiny-id/checkpoint-900",
+  "epoch": 6.064,
   "eval_steps": 100,
-  "global_step": 500,
+  "global_step": 1000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -11,7 +11,7 @@
     {
       "epoch": 0.03,
       "learning_rate": 4.898977360288234e-06,
-      "loss": 1.7847,
+      "loss": 1.7848,
       "step": 25
     },
     {
@@ -23,7 +23,7 @@
     {
       "epoch": 0.07,
       "learning_rate": 6.85912902234906e-06,
-      "loss": 1.1407,
+      "loss": 1.1406,
       "step": 75
     },
     {
@@ -35,10 +35,10 @@
     {
       "epoch": 0.1,
       "eval_loss": 0.8486328125,
-      "eval_runtime": 1294.216,
-      "eval_samples_per_second": 2.814,
-      "eval_steps_per_second": 0.176,
-      "eval_wer": 54.17241696568221,
+      "eval_runtime": 1287.4485,
+      "eval_samples_per_second": 2.829,
+      "eval_steps_per_second": 0.177,
+      "eval_wer": 54.24602079308124,
       "step": 100
     },
     {
@@ -68,22 +68,22 @@
     {
       "epoch": 1.04,
       "eval_loss": 0.7578125,
-      "eval_runtime": 1161.7981,
-      "eval_samples_per_second": 3.135,
-      "eval_steps_per_second": 0.196,
-      "eval_wer": 48.339313644309506,
+      "eval_runtime": 1159.7848,
+      "eval_samples_per_second": 3.14,
+      "eval_steps_per_second": 0.197,
+      "eval_wer": 48.389916275646335,
       "step": 200
     },
     {
       "epoch": 1.07,
       "learning_rate": 8.686247975778677e-06,
-      "loss": 0.8413,
+      "loss": 0.8414,
       "step": 225
     },
     {
       "epoch": 1.09,
       "learning_rate": 8.858694625217149e-06,
-      "loss": 0.68,
+      "loss": 0.6799,
       "step": 250
     },
     {
@@ -101,22 +101,22 @@
     {
       "epoch": 1.14,
       "eval_loss": 0.73876953125,
-      "eval_runtime": 1191.6636,
-      "eval_samples_per_second": 3.056,
-      "eval_steps_per_second": 0.191,
-      "eval_wer": 49.25936148679732,
+      "eval_runtime": 1190.5006,
+      "eval_samples_per_second": 3.059,
+      "eval_steps_per_second": 0.192,
+      "eval_wer": 49.3283650749839,
       "step": 300
     },
     {
       "epoch": 2.01,
       "learning_rate": 9.28689473531776e-06,
-      "loss": 0.5102,
+      "loss": 0.5103,
       "step": 325
     },
     {
       "epoch": 2.04,
       "learning_rate": 9.407574351377137e-06,
-      "loss": 0.602,
+      "loss": 0.6019,
       "step": 350
     },
     {
@@ -134,10 +134,10 @@
     {
       "epoch": 2.09,
       "eval_loss": 0.7177734375,
-      "eval_runtime": 1169.8926,
-      "eval_samples_per_second": 3.113,
-      "eval_steps_per_second": 0.195,
-      "eval_wer": 46.126598583126324,
+      "eval_runtime": 1156.6175,
+      "eval_samples_per_second": 3.149,
+      "eval_steps_per_second": 0.197,
+      "eval_wer": 45.795381359830714,
       "step": 400
     },
     {
@@ -161,34 +161,199 @@
     {
       "epoch": 3.03,
       "learning_rate": 9.987075336738768e-06,
-      "loss": 0.476,
+      "loss": 0.4761,
       "step": 500
     },
     {
       "epoch": 3.03,
       "eval_loss": 0.7109375,
-      "eval_runtime": 1126.9534,
-      "eval_samples_per_second": 3.232,
+      "eval_runtime": 1128.5074,
+      "eval_samples_per_second": 3.227,
       "eval_steps_per_second": 0.202,
-      "eval_wer": 45.243352654338025,
+      "eval_wer": 45.21575121906339,
       "step": 500
     },
     {
-      "epoch": 3.03,
-      "step": 500,
-      "total_flos": 3.922034448169697e+17,
-      "train_loss": 0.0,
-      "train_runtime": 67.5132,
-      "train_samples_per_second": 947.963,
-      "train_steps_per_second": 29.624
+      "epoch": 3.06,
+      "learning_rate": 9.600000000000001e-06,
+      "loss": 0.4891,
+      "step": 525
+    },
+    {
+      "epoch": 3.08,
+      "learning_rate": 9.100000000000001e-06,
+      "loss": 0.4411,
+      "step": 550
+    },
+    {
+      "epoch": 3.11,
+      "learning_rate": 8.6e-06,
+      "loss": 0.367,
+      "step": 575
+    },
+    {
+      "epoch": 3.13,
+      "learning_rate": 8.1e-06,
+      "loss": 0.2674,
+      "step": 600
+    },
+    {
+      "epoch": 3.13,
+      "eval_loss": 0.70068359375,
+      "eval_runtime": 1151.1688,
+      "eval_samples_per_second": 3.164,
+      "eval_steps_per_second": 0.198,
+      "eval_wer": 44.843131842855826,
+      "step": 600
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 7.600000000000001e-06,
+      "loss": 0.2263,
+      "step": 625
+    },
+    {
+      "epoch": 4.03,
+      "learning_rate": 7.100000000000001e-06,
+      "loss": 0.3527,
+      "step": 650
+    },
+    {
+      "epoch": 4.05,
+      "learning_rate": 6.600000000000001e-06,
+      "loss": 0.369,
+      "step": 675
+    },
+    {
+      "epoch": 4.08,
+      "learning_rate": 6.1e-06,
+      "loss": 0.3628,
+      "step": 700
+    },
+    {
+      "epoch": 4.08,
+      "eval_loss": 0.70263671875,
+      "eval_runtime": 1168.9802,
+      "eval_samples_per_second": 3.116,
+      "eval_steps_per_second": 0.195,
+      "eval_wer": 44.24970098445119,
+      "step": 700
+    },
+    {
+      "epoch": 4.1,
+      "learning_rate": 5.600000000000001e-06,
+      "loss": 0.2981,
+      "step": 725
+    },
+    {
+      "epoch": 4.13,
+      "learning_rate": 5.1e-06,
+      "loss": 0.2296,
+      "step": 750
+    },
+    {
+      "epoch": 4.15,
+      "learning_rate": 4.600000000000001e-06,
+      "loss": 0.1694,
+      "step": 775
+    },
+    {
+      "epoch": 5.02,
+      "learning_rate": 4.1e-06,
+      "loss": 0.2565,
+      "step": 800
+    },
+    {
+      "epoch": 5.02,
+      "eval_loss": 0.70849609375,
+      "eval_runtime": 1151.0035,
+      "eval_samples_per_second": 3.164,
+      "eval_steps_per_second": 0.198,
+      "eval_wer": 44.50731438034778,
+      "step": 800
+    },
+    {
+      "epoch": 5.04,
+      "learning_rate": 3.6000000000000003e-06,
+      "loss": 0.2965,
+      "step": 825
+    },
+    {
+      "epoch": 5.07,
+      "learning_rate": 3.1000000000000004e-06,
+      "loss": 0.2987,
+      "step": 850
+    },
+    {
+      "epoch": 5.09,
+      "learning_rate": 2.6e-06,
+      "loss": 0.2593,
+      "step": 875
+    },
+    {
+      "epoch": 5.12,
+      "learning_rate": 2.1000000000000002e-06,
+      "loss": 0.2147,
+      "step": 900
+    },
+    {
+      "epoch": 5.12,
+      "eval_loss": 0.708984375,
+      "eval_runtime": 1111.0874,
+      "eval_samples_per_second": 3.278,
+      "eval_steps_per_second": 0.205,
+      "eval_wer": 43.44465912227436,
+      "step": 900
+    },
+    {
+      "epoch": 5.14,
+      "learning_rate": 1.6000000000000001e-06,
+      "loss": 0.1495,
+      "step": 925
+    },
+    {
+      "epoch": 6.01,
+      "learning_rate": 1.1e-06,
+      "loss": 0.1969,
+      "step": 950
+    },
+    {
+      "epoch": 6.04,
+      "learning_rate": 6.000000000000001e-07,
+      "loss": 0.2518,
+      "step": 975
+    },
+    {
+      "epoch": 6.06,
+      "learning_rate": 1.0000000000000001e-07,
+      "loss": 0.28,
+      "step": 1000
+    },
+    {
+      "epoch": 6.06,
+      "eval_loss": 0.71337890625,
+      "eval_runtime": 1142.0113,
+      "eval_samples_per_second": 3.189,
+      "eval_steps_per_second": 0.2,
+      "eval_wer": 43.95528567485509,
+      "step": 1000
+    },
+    {
+      "epoch": 6.06,
+      "step": 1000,
+      "total_flos": 7.844068896339395e+17,
+      "train_loss": 0.5104879150390625,
+      "train_runtime": 24718.0782,
+      "train_samples_per_second": 1.295,
+      "train_steps_per_second": 0.04
     }
   ],
   "logging_steps": 25,
-  "max_steps": 2000,
+  "max_steps": 1000,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 9223372036854775807,
   "save_steps": 100,
-  "total_flos": 3.922034448169697e+17,
+  "total_flos": 7.844068896339395e+17,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": null