End of training

README.md

@@ -1,7 +1,12 @@
 ---
+language:
+- hi
 license: cc-by-nc-4.0
 base_model: facebook/mms-1b-all
 tags:
+- automatic-speech-recognition
+- mozilla-foundation/common_voice_16_0
+- mms
 - generated_from_trainer
 datasets:
 - common_voice_16_0
@@ -14,15 +19,15 @@ model-index:
       name: Automatic Speech Recognition
       type: automatic-speech-recognition
     dataset:
-      name: common_voice_16_0
+      name: MOZILLA-FOUNDATION/COMMON_VOICE_16_0 - HI
       type: common_voice_16_0
       config: hi
       split: test
-      args: hi
+      args: 'Config: hi, Training split: train+validation, Eval split: test'
     metrics:
     - name: Wer
       type: wer
-      value: 0.2522266734082688
+      value: 0.2516432655283731
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -30,10 +35,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # wav2vec2-common_voice-hi-mms-demo
 
-This model is a fine-tuned version of [facebook/mms-1b-all](https://huggingface.co/facebook/mms-1b-all) on the common_voice_16_0 dataset.
+This model is a fine-tuned version of [facebook/mms-1b-all](https://huggingface.co/facebook/mms-1b-all) on the MOZILLA-FOUNDATION/COMMON_VOICE_16_0 - HI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.2673
-- Wer: 0.2522
+- Loss: 0.2672
+- Wer: 0.2516
 
 ## Model description
 

adapter.hin.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e23f3d274deb915f80719d66791a0026a6214af145cea237ac7a044f816a2593
+size 9162368

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 4.0,
+    "eval_loss": 0.26723530888557434,
+    "eval_runtime": 757.5485,
+    "eval_samples": 3107,
+    "eval_samples_per_second": 4.101,
+    "eval_steps_per_second": 0.513,
+    "eval_wer": 0.2516432655283731,
+    "train_loss": 0.471761938411549,
+    "train_runtime": 38765.3056,
+    "train_samples": 7084,
+    "train_samples_per_second": 0.731,
+    "train_steps_per_second": 0.091
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 4.0,
+    "eval_loss": 0.26723530888557434,
+    "eval_runtime": 757.5485,
+    "eval_samples": 3107,
+    "eval_samples_per_second": 4.101,
+    "eval_steps_per_second": 0.513,
+    "eval_wer": 0.2516432655283731
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 4.0,
+    "train_loss": 0.471761938411549,
+    "train_runtime": 38765.3056,
+    "train_samples": 7084,
+    "train_samples_per_second": 0.731,
+    "train_steps_per_second": 0.091
+}

trainer_state.json

@@ -0,0 +1,387 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 4.0,
+  "eval_steps": 100,
+  "global_step": 3544,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.11,
+      "eval_loss": 0.4486841857433319,
+      "eval_runtime": 759.5602,
+      "eval_samples_per_second": 4.091,
+      "eval_steps_per_second": 0.512,
+      "eval_wer": 0.35654000233363153,
+      "step": 100
+    },
+    {
+      "epoch": 0.23,
+      "eval_loss": 0.3544192612171173,
+      "eval_runtime": 754.7374,
+      "eval_samples_per_second": 4.117,
+      "eval_steps_per_second": 0.515,
+      "eval_wer": 0.33168682665007193,
+      "step": 200
+    },
+    {
+      "epoch": 0.34,
+      "eval_loss": 0.36927035450935364,
+      "eval_runtime": 750.419,
+      "eval_samples_per_second": 4.14,
+      "eval_steps_per_second": 0.518,
+      "eval_wer": 0.308817237758158,
+      "step": 300
+    },
+    {
+      "epoch": 0.45,
+      "eval_loss": 0.3404374122619629,
+      "eval_runtime": 754.9059,
+      "eval_samples_per_second": 4.116,
+      "eval_steps_per_second": 0.515,
+      "eval_wer": 0.3040332931430127,
+      "step": 400
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 0.0008853077816492451,
+      "loss": 1.5084,
+      "step": 500
+    },
+    {
+      "epoch": 0.56,
+      "eval_loss": 0.33462777733802795,
+      "eval_runtime": 756.1347,
+      "eval_samples_per_second": 4.109,
+      "eval_steps_per_second": 0.514,
+      "eval_wer": 0.29952160553848545,
+      "step": 500
+    },
+    {
+      "epoch": 0.68,
+      "eval_loss": 0.34105485677719116,
+      "eval_runtime": 756.3042,
+      "eval_samples_per_second": 4.108,
+      "eval_steps_per_second": 0.514,
+      "eval_wer": 0.2935708451635487,
+      "step": 600
+    },
+    {
+      "epoch": 0.79,
+      "eval_loss": 0.31745076179504395,
+      "eval_runtime": 748.8702,
+      "eval_samples_per_second": 4.149,
+      "eval_steps_per_second": 0.519,
+      "eval_wer": 0.2887480066897437,
+      "step": 700
+    },
+    {
+      "epoch": 0.9,
+      "eval_loss": 0.3159240484237671,
+      "eval_runtime": 765.2091,
+      "eval_samples_per_second": 4.06,
+      "eval_steps_per_second": 0.508,
+      "eval_wer": 0.2898370347322158,
+      "step": 800
+    },
+    {
+      "epoch": 1.02,
+      "eval_loss": 0.3138948380947113,
+      "eval_runtime": 752.8617,
+      "eval_samples_per_second": 4.127,
+      "eval_steps_per_second": 0.517,
+      "eval_wer": 0.304538913305589,
+      "step": 900
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 0.000740418118466899,
+      "loss": 0.3485,
+      "step": 1000
+    },
+    {
+      "epoch": 1.13,
+      "eval_loss": 0.3067200779914856,
+      "eval_runtime": 749.2127,
+      "eval_samples_per_second": 4.147,
+      "eval_steps_per_second": 0.519,
+      "eval_wer": 0.2957877951071526,
+      "step": 1000
+    },
+    {
+      "epoch": 1.24,
+      "eval_loss": 0.29693612456321716,
+      "eval_runtime": 752.9596,
+      "eval_samples_per_second": 4.126,
+      "eval_steps_per_second": 0.517,
+      "eval_wer": 0.2766909105052312,
+      "step": 1100
+    },
+    {
+      "epoch": 1.35,
+      "eval_loss": 0.29161983728408813,
+      "eval_runtime": 757.6219,
+      "eval_samples_per_second": 4.101,
+      "eval_steps_per_second": 0.513,
+      "eval_wer": 0.2714013457275096,
+      "step": 1200
+    },
+    {
+      "epoch": 1.47,
+      "eval_loss": 0.2893225848674774,
+      "eval_runtime": 751.196,
+      "eval_samples_per_second": 4.136,
+      "eval_steps_per_second": 0.518,
+      "eval_wer": 0.2663451441017463,
+      "step": 1300
+    },
+    {
+      "epoch": 1.58,
+      "eval_loss": 0.3183298408985138,
+      "eval_runtime": 752.5817,
+      "eval_samples_per_second": 4.128,
+      "eval_steps_per_second": 0.517,
+      "eval_wer": 0.29854925907199253,
+      "step": 1400
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 0.0005952380952380953,
+      "loss": 0.3152,
+      "step": 1500
+    },
+    {
+      "epoch": 1.69,
+      "eval_loss": 0.296146422624588,
+      "eval_runtime": 747.739,
+      "eval_samples_per_second": 4.155,
+      "eval_steps_per_second": 0.52,
+      "eval_wer": 0.26875656333864884,
+      "step": 1500
+    },
+    {
+      "epoch": 1.81,
+      "eval_loss": 0.2847990095615387,
+      "eval_runtime": 745.157,
+      "eval_samples_per_second": 4.17,
+      "eval_steps_per_second": 0.522,
+      "eval_wer": 0.26653961339504495,
+      "step": 1600
+    },
+    {
+      "epoch": 1.92,
+      "eval_loss": 0.28440287709236145,
+      "eval_runtime": 753.6747,
+      "eval_samples_per_second": 4.122,
+      "eval_steps_per_second": 0.516,
+      "eval_wer": 0.2656450546458714,
+      "step": 1700
+    },
+    {
+      "epoch": 2.03,
+      "eval_loss": 0.2854562997817993,
+      "eval_runtime": 748.224,
+      "eval_samples_per_second": 4.153,
+      "eval_steps_per_second": 0.52,
+      "eval_wer": 0.2707401501302944,
+      "step": 1800
+    },
+    {
+      "epoch": 2.14,
+      "eval_loss": 0.2886996865272522,
+      "eval_runtime": 750.8758,
+      "eval_samples_per_second": 4.138,
+      "eval_steps_per_second": 0.518,
+      "eval_wer": 0.26860098790400994,
+      "step": 1900
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 0.0004500580720092916,
+      "loss": 0.3058,
+      "step": 2000
+    },
+    {
+      "epoch": 2.26,
+      "eval_loss": 0.28578099608421326,
+      "eval_runtime": 748.0813,
+      "eval_samples_per_second": 4.153,
+      "eval_steps_per_second": 0.52,
+      "eval_wer": 0.26568394850453114,
+      "step": 2000
+    },
+    {
+      "epoch": 2.37,
+      "eval_loss": 0.2813587188720703,
+      "eval_runtime": 756.7359,
+      "eval_samples_per_second": 4.106,
+      "eval_steps_per_second": 0.514,
+      "eval_wer": 0.2629224845396912,
+      "step": 2100
+    },
+    {
+      "epoch": 2.48,
+      "eval_loss": 0.2809281349182129,
+      "eval_runtime": 759.1954,
+      "eval_samples_per_second": 4.092,
+      "eval_steps_per_second": 0.512,
+      "eval_wer": 0.26331142312628836,
+      "step": 2200
+    },
+    {
+      "epoch": 2.6,
+      "eval_loss": 0.2778892517089844,
+      "eval_runtime": 756.4055,
+      "eval_samples_per_second": 4.108,
+      "eval_steps_per_second": 0.514,
+      "eval_wer": 0.26128894247598305,
+      "step": 2300
+    },
+    {
+      "epoch": 2.71,
+      "eval_loss": 0.2744755446910858,
+      "eval_runtime": 758.3998,
+      "eval_samples_per_second": 4.097,
+      "eval_steps_per_second": 0.513,
+      "eval_wer": 0.25806075220722646,
+      "step": 2400
+    },
+    {
+      "epoch": 2.82,
+      "learning_rate": 0.0003051684088269454,
+      "loss": 0.2861,
+      "step": 2500
+    },
+    {
+      "epoch": 2.82,
+      "eval_loss": 0.2769048810005188,
+      "eval_runtime": 757.4274,
+      "eval_samples_per_second": 4.102,
+      "eval_steps_per_second": 0.514,
+      "eval_wer": 0.26175566877989964,
+      "step": 2500
+    },
+    {
+      "epoch": 2.93,
+      "eval_loss": 0.2742438316345215,
+      "eval_runtime": 761.8109,
+      "eval_samples_per_second": 4.078,
+      "eval_steps_per_second": 0.511,
+      "eval_wer": 0.2575940259033099,
+      "step": 2600
+    },
+    {
+      "epoch": 3.05,
+      "eval_loss": 0.27301648259162903,
+      "eval_runtime": 765.5965,
+      "eval_samples_per_second": 4.058,
+      "eval_steps_per_second": 0.508,
+      "eval_wer": 0.2574773443273307,
+      "step": 2700
+    },
+    {
+      "epoch": 3.16,
+      "eval_loss": 0.27274471521377563,
+      "eval_runtime": 756.2486,
+      "eval_samples_per_second": 4.108,
+      "eval_steps_per_second": 0.514,
+      "eval_wer": 0.25638831628485864,
+      "step": 2800
+    },
+    {
+      "epoch": 3.27,
+      "eval_loss": 0.27257227897644043,
+      "eval_runtime": 762.5719,
+      "eval_samples_per_second": 4.074,
+      "eval_steps_per_second": 0.51,
+      "eval_wer": 0.2562716347088795,
+      "step": 2900
+    },
+    {
+      "epoch": 3.39,
+      "learning_rate": 0.0001599883855981417,
+      "loss": 0.2839,
+      "step": 3000
+    },
+    {
+      "epoch": 3.39,
+      "eval_loss": 0.2713397443294525,
+      "eval_runtime": 762.4867,
+      "eval_samples_per_second": 4.075,
+      "eval_steps_per_second": 0.51,
+      "eval_wer": 0.25755513204465014,
+      "step": 3000
+    },
+    {
+      "epoch": 3.5,
+      "eval_loss": 0.26895028352737427,
+      "eval_runtime": 763.0484,
+      "eval_samples_per_second": 4.072,
+      "eval_steps_per_second": 0.51,
+      "eval_wer": 0.2536657461786784,
+      "step": 3100
+    },
+    {
+      "epoch": 3.61,
+      "eval_loss": 0.2706141769886017,
+      "eval_runtime": 754.9996,
+      "eval_samples_per_second": 4.115,
+      "eval_steps_per_second": 0.515,
+      "eval_wer": 0.2539768970479561,
+      "step": 3200
+    },
+    {
+      "epoch": 3.72,
+      "eval_loss": 0.26873159408569336,
+      "eval_runtime": 760.416,
+      "eval_samples_per_second": 4.086,
+      "eval_steps_per_second": 0.512,
+      "eval_wer": 0.25421026019991444,
+      "step": 3300
+    },
+    {
+      "epoch": 3.84,
+      "eval_loss": 0.26714619994163513,
+      "eval_runtime": 754.0393,
+      "eval_samples_per_second": 4.12,
+      "eval_steps_per_second": 0.516,
+      "eval_wer": 0.2521488856909494,
+      "step": 3400
+    },
+    {
+      "epoch": 3.95,
+      "learning_rate": 1.5098722415795587e-05,
+      "loss": 0.2706,
+      "step": 3500
+    },
+    {
+      "epoch": 3.95,
+      "eval_loss": 0.267267644405365,
+      "eval_runtime": 758.6405,
+      "eval_samples_per_second": 4.095,
+      "eval_steps_per_second": 0.513,
+      "eval_wer": 0.2522266734082688,
+      "step": 3500
+    },
+    {
+      "epoch": 4.0,
+      "step": 3544,
+      "total_flos": 1.1766650360089125e+19,
+      "train_loss": 0.471761938411549,
+      "train_runtime": 38765.3056,
+      "train_samples_per_second": 0.731,
+      "train_steps_per_second": 0.091
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 3544,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 4,
+  "save_steps": 200,
+  "total_flos": 1.1766650360089125e+19,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}