End of training

README.md

@@ -1,39 +1,42 @@
 ---
+language:
+- te
 license: apache-2.0
 base_model: openai/whisper-base
 tags:
+- whisper-event
 - generated_from_trainer
 datasets:
-- common_voice_16_0
+- mozilla-foundation/common_voice_16_0
 metrics:
 - wer
 model-index:
-- name: breeze-listen-dsw-base-te
+- name: Breeze DSW Telugu - base
   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 te
+      type: mozilla-foundation/common_voice_16_0
       config: te
       split: test
       args: te
     metrics:
     - name: Wer
       type: wer
-      value: 90.65934065934066
+      value: 89.28571428571429
 ---
 
 <!-- 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-listen-dsw-base-te
+# Breeze DSW Telugu - base
 
-This model is a fine-tuned version of [openai/whisper-base](https://huggingface.co/openai/whisper-base) on the common_voice_16_0 dataset.
+This model is a fine-tuned version of [openai/whisper-base](https://huggingface.co/openai/whisper-base) on the mozilla-foundation/common_voice_16_0 te dataset.
 It achieves the following results on the evaluation set:
-- Loss: 2.2441
-- Wer: 90.6593
+- Loss: 2.0469
+- Wer: 89.2857
 
 ## Model description
 

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 199.0,
+    "eval_loss": 2.046875,
+    "eval_runtime": 53.1669,
+    "eval_samples_per_second": 0.508,
+    "eval_steps_per_second": 0.038,
+    "eval_wer": 89.28571428571429,
+    "train_loss": 0.0,
+    "train_runtime": 16.6622,
+    "train_samples_per_second": 3841.032,
+    "train_steps_per_second": 120.032
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 199.0,
+    "eval_loss": 2.046875,
+    "eval_runtime": 53.1669,
+    "eval_samples_per_second": 0.508,
+    "eval_steps_per_second": 0.038,
+    "eval_wer": 89.28571428571429
+}

runs/Jan18_11-19-00_knight/events.out.tfevents.1705573270.knight.157497.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8241354c2c5f8e8862d9e697ecc821956d30e5ea83b0355efa9cb2e92e683f8d
+size 406

train_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 199.0,
+    "train_loss": 0.0,
+    "train_runtime": 16.6622,
+    "train_samples_per_second": 3841.032,
+    "train_steps_per_second": 120.032
+}

trainer_state.json

@@ -0,0 +1,144 @@
+{
+  "best_metric": 89.28571428571429,
+  "best_model_checkpoint": "/cosmos/home/sp-operator/ai/training/models/huggingface/scripts/../breeze-listen-dsw-base-te/checkpoint-200",
+  "epoch": 199.001,
+  "eval_steps": 200,
+  "global_step": 400,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 12.0,
+      "learning_rate": 4.973833272194737e-06,
+      "loss": 1.7314,
+      "step": 25
+    },
+    {
+      "epoch": 24.0,
+      "learning_rate": 6.195318418690893e-06,
+      "loss": 1.0852,
+      "step": 50
+    },
+    {
+      "epoch": 37.0,
+      "learning_rate": 6.881634451095711e-06,
+      "loss": 0.4273,
+      "step": 75
+    },
+    {
+      "epoch": 49.0,
+      "learning_rate": 7.361221988663844e-06,
+      "loss": 0.0907,
+      "step": 100
+    },
+    {
+      "epoch": 62.0,
+      "learning_rate": 7.730207550743121e-06,
+      "loss": 0.0175,
+      "step": 125
+    },
+    {
+      "epoch": 74.0,
+      "learning_rate": 8.03016458599496e-06,
+      "loss": 0.0032,
+      "step": 150
+    },
+    {
+      "epoch": 87.0,
+      "learning_rate": 8.282894746203441e-06,
+      "loss": 0.0011,
+      "step": 175
+    },
+    {
+      "epoch": 99.0,
+      "learning_rate": 8.501266121799902e-06,
+      "loss": 0.0007,
+      "step": 200
+    },
+    {
+      "epoch": 99.0,
+      "eval_loss": 2.046875,
+      "eval_runtime": 63.8247,
+      "eval_samples_per_second": 0.423,
+      "eval_steps_per_second": 0.031,
+      "eval_wer": 89.28571428571429,
+      "step": 200
+    },
+    {
+      "epoch": 112.0,
+      "learning_rate": 8.693512601774437e-06,
+      "loss": 0.0005,
+      "step": 225
+    },
+    {
+      "epoch": 124.0,
+      "learning_rate": 8.865222471593567e-06,
+      "loss": 0.0004,
+      "step": 250
+    },
+    {
+      "epoch": 137.0,
+      "learning_rate": 9.020362953730323e-06,
+      "loss": 0.0003,
+      "step": 275
+    },
+    {
+      "epoch": 149.0,
+      "learning_rate": 9.161852281961698e-06,
+      "loss": 0.0003,
+      "step": 300
+    },
+    {
+      "epoch": 162.0,
+      "learning_rate": 9.29189975311636e-06,
+      "loss": 0.0002,
+      "step": 325
+    },
+    {
+      "epoch": 174.0,
+      "learning_rate": 9.412218256259678e-06,
+      "loss": 0.0002,
+      "step": 350
+    },
+    {
+      "epoch": 187.0,
+      "learning_rate": 9.524162683365145e-06,
+      "loss": 0.0002,
+      "step": 375
+    },
+    {
+      "epoch": 199.0,
+      "learning_rate": 9.62882322733502e-06,
+      "loss": 0.0002,
+      "step": 400
+    },
+    {
+      "epoch": 199.0,
+      "eval_loss": 2.244140625,
+      "eval_runtime": 56.1817,
+      "eval_samples_per_second": 0.481,
+      "eval_steps_per_second": 0.036,
+      "eval_wer": 90.65934065934066,
+      "step": 400
+    },
+    {
+      "epoch": 199.0,
+      "step": 400,
+      "total_flos": 5.059079773670605e+17,
+      "train_loss": 0.0,
+      "train_runtime": 16.6622,
+      "train_samples_per_second": 3841.032,
+      "train_steps_per_second": 120.032
+    }
+  ],
+  "logging_steps": 25,
+  "max_steps": 2000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 9223372036854775807,
+  "save_steps": 200,
+  "total_flos": 5.059079773670605e+17,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}