End of training

README.md

@@ -1,7 +1,12 @@
 ---
+language:
+- id
 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 - ID
       type: common_voice_16_0
       config: id
       split: test
-      args: id
+      args: 'Config: id, Training split: train+validation, Eval split: test'
     metrics:
     - name: Wer
       type: wer
-      value: 0.1455760839290688
+      value: 0.145808188654721
 ---
 
 <!-- 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. -->
 
 # breeze-listen-w2v2-id
 
-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 - ID dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.1253
-- Wer: 0.1456
+- Wer: 0.1458
 
 ## Model description
 

adapter.ind.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fabda2a21f8bfe6ecaf473ac7ed25810dbdf35a63ae88262d894ebb86c779902
+size 8860028

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 4.0,
+    "eval_loss": 0.12526944279670715,
+    "eval_runtime": 688.3498,
+    "eval_samples": 3642,
+    "eval_samples_per_second": 5.291,
+    "eval_steps_per_second": 0.662,
+    "eval_wer": 0.145808188654721,
+    "train_loss": 0.7944976037459608,
+    "train_runtime": 46739.0877,
+    "train_samples": 8309,
+    "train_samples_per_second": 0.711,
+    "train_steps_per_second": 0.178
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 4.0,
+    "eval_loss": 0.12526944279670715,
+    "eval_runtime": 688.3498,
+    "eval_samples": 3642,
+    "eval_samples_per_second": 5.291,
+    "eval_steps_per_second": 0.662,
+    "eval_wer": 0.145808188654721
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 4.0,
+    "train_loss": 0.7944976037459608,
+    "train_runtime": 46739.0877,
+    "train_samples": 8309,
+    "train_samples_per_second": 0.711,
+    "train_steps_per_second": 0.178
+}

trainer_state.json

@@ -0,0 +1,495 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 4.0,
+  "eval_steps": 200,
+  "global_step": 8312,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.1,
+      "eval_loss": 3.2670648097991943,
+      "eval_runtime": 692.0326,
+      "eval_samples_per_second": 5.263,
+      "eval_steps_per_second": 0.659,
+      "eval_wer": 1.0,
+      "step": 200
+    },
+    {
+      "epoch": 0.19,
+      "eval_loss": 2.8740603923797607,
+      "eval_runtime": 683.8046,
+      "eval_samples_per_second": 5.326,
+      "eval_steps_per_second": 0.667,
+      "eval_wer": 1.0006963141769567,
+      "step": 400
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 0.0009517778860204579,
+      "loss": 3.8381,
+      "step": 500
+    },
+    {
+      "epoch": 0.29,
+      "eval_loss": 2.761221170425415,
+      "eval_runtime": 683.642,
+      "eval_samples_per_second": 5.327,
+      "eval_steps_per_second": 0.667,
+      "eval_wer": 0.9954507473772166,
+      "step": 600
+    },
+    {
+      "epoch": 0.38,
+      "eval_loss": 2.633348226547241,
+      "eval_runtime": 684.1815,
+      "eval_samples_per_second": 5.323,
+      "eval_steps_per_second": 0.666,
+      "eval_wer": 0.9981431621947823,
+      "step": 800
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 0.000890891378470531,
+      "loss": 2.6996,
+      "step": 1000
+    },
+    {
+      "epoch": 0.48,
+      "eval_loss": 2.3073549270629883,
+      "eval_runtime": 686.3923,
+      "eval_samples_per_second": 5.306,
+      "eval_steps_per_second": 0.664,
+      "eval_wer": 0.9770680531055612,
+      "step": 1000
+    },
+    {
+      "epoch": 0.58,
+      "eval_loss": 2.0154612064361572,
+      "eval_runtime": 686.5478,
+      "eval_samples_per_second": 5.305,
+      "eval_steps_per_second": 0.664,
+      "eval_wer": 0.9286045863893789,
+      "step": 1200
+    },
+    {
+      "epoch": 0.67,
+      "eval_loss": 1.9155136346817017,
+      "eval_runtime": 689.547,
+      "eval_samples_per_second": 5.282,
+      "eval_steps_per_second": 0.661,
+      "eval_wer": 0.8947172964441557,
+      "step": 1400
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 0.000830004870920604,
+      "loss": 2.2919,
+      "step": 1500
+    },
+    {
+      "epoch": 0.77,
+      "eval_loss": 1.641204595565796,
+      "eval_runtime": 685.3748,
+      "eval_samples_per_second": 5.314,
+      "eval_steps_per_second": 0.665,
+      "eval_wer": 0.8813944851917185,
+      "step": 1600
+    },
+    {
+      "epoch": 0.87,
+      "eval_loss": 1.4531193971633911,
+      "eval_runtime": 689.2035,
+      "eval_samples_per_second": 5.284,
+      "eval_steps_per_second": 0.662,
+      "eval_wer": 0.8285210286881441,
+      "step": 1800
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 0.0007691183633706771,
+      "loss": 1.5872,
+      "step": 2000
+    },
+    {
+      "epoch": 0.96,
+      "eval_loss": 0.1812867820262909,
+      "eval_runtime": 685.9058,
+      "eval_samples_per_second": 5.31,
+      "eval_steps_per_second": 0.665,
+      "eval_wer": 0.2060161544889054,
+      "step": 2000
+    },
+    {
+      "epoch": 1.06,
+      "eval_loss": 0.1635832041501999,
+      "eval_runtime": 687.9409,
+      "eval_samples_per_second": 5.294,
+      "eval_steps_per_second": 0.663,
+      "eval_wer": 0.18062389750255314,
+      "step": 2200
+    },
+    {
+      "epoch": 1.15,
+      "eval_loss": 0.155806764960289,
+      "eval_runtime": 692.1735,
+      "eval_samples_per_second": 5.262,
+      "eval_steps_per_second": 0.659,
+      "eval_wer": 0.17444991180020425,
+      "step": 2400
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 0.0007084754018509498,
+      "loss": 0.2659,
+      "step": 2500
+    },
+    {
+      "epoch": 1.25,
+      "eval_loss": 0.152183398604393,
+      "eval_runtime": 688.556,
+      "eval_samples_per_second": 5.289,
+      "eval_steps_per_second": 0.662,
+      "eval_wer": 0.1646550923776808,
+      "step": 2600
+    },
+    {
+      "epoch": 1.35,
+      "eval_loss": 0.15532232820987701,
+      "eval_runtime": 688.1144,
+      "eval_samples_per_second": 5.293,
+      "eval_steps_per_second": 0.663,
+      "eval_wer": 0.16641908829263763,
+      "step": 2800
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 0.0006475888943010228,
+      "loss": 0.2436,
+      "step": 3000
+    },
+    {
+      "epoch": 1.44,
+      "eval_loss": 0.1840931922197342,
+      "eval_runtime": 692.811,
+      "eval_samples_per_second": 5.257,
+      "eval_steps_per_second": 0.658,
+      "eval_wer": 0.1960820722309906,
+      "step": 3000
+    },
+    {
+      "epoch": 1.54,
+      "eval_loss": 0.14190182089805603,
+      "eval_runtime": 690.3365,
+      "eval_samples_per_second": 5.276,
+      "eval_steps_per_second": 0.661,
+      "eval_wer": 0.1640051991458546,
+      "step": 3200
+    },
+    {
+      "epoch": 1.64,
+      "eval_loss": 0.14559713006019592,
+      "eval_runtime": 685.0999,
+      "eval_samples_per_second": 5.316,
+      "eval_steps_per_second": 0.666,
+      "eval_wer": 0.17143255036672547,
+      "step": 3400
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 0.000586702386751096,
+      "loss": 0.2464,
+      "step": 3500
+    },
+    {
+      "epoch": 1.73,
+      "eval_loss": 0.14024095237255096,
+      "eval_runtime": 692.5402,
+      "eval_samples_per_second": 5.259,
+      "eval_steps_per_second": 0.658,
+      "eval_wer": 0.16070931204159317,
+      "step": 3600
+    },
+    {
+      "epoch": 1.83,
+      "eval_loss": 0.1345185786485672,
+      "eval_runtime": 694.4502,
+      "eval_samples_per_second": 5.244,
+      "eval_steps_per_second": 0.657,
+      "eval_wer": 0.1528177513694179,
+      "step": 3800
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 0.000525815879201169,
+      "loss": 0.2292,
+      "step": 4000
+    },
+    {
+      "epoch": 1.92,
+      "eval_loss": 0.134234219789505,
+      "eval_runtime": 693.8578,
+      "eval_samples_per_second": 5.249,
+      "eval_steps_per_second": 0.657,
+      "eval_wer": 0.155556587132114,
+      "step": 4000
+    },
+    {
+      "epoch": 2.02,
+      "eval_loss": 0.13340923190116882,
+      "eval_runtime": 684.0209,
+      "eval_samples_per_second": 5.324,
+      "eval_steps_per_second": 0.667,
+      "eval_wer": 0.15518521957107045,
+      "step": 4200
+    },
+    {
+      "epoch": 2.12,
+      "eval_loss": 0.13518257439136505,
+      "eval_runtime": 687.2622,
+      "eval_samples_per_second": 5.299,
+      "eval_steps_per_second": 0.664,
+      "eval_wer": 0.1543496425587225,
+      "step": 4400
+    },
+    {
+      "epoch": 2.17,
+      "learning_rate": 0.0004649293716512421,
+      "loss": 0.2209,
+      "step": 4500
+    },
+    {
+      "epoch": 2.21,
+      "eval_loss": 0.13499902188777924,
+      "eval_runtime": 691.8679,
+      "eval_samples_per_second": 5.264,
+      "eval_steps_per_second": 0.659,
+      "eval_wer": 0.1537925912171572,
+      "step": 4600
+    },
+    {
+      "epoch": 2.31,
+      "eval_loss": 0.13418444991111755,
+      "eval_runtime": 685.3615,
+      "eval_samples_per_second": 5.314,
+      "eval_steps_per_second": 0.665,
+      "eval_wer": 0.1530498560950701,
+      "step": 4800
+    },
+    {
+      "epoch": 2.41,
+      "learning_rate": 0.00040404286410131515,
+      "loss": 0.2136,
+      "step": 5000
+    },
+    {
+      "epoch": 2.41,
+      "eval_loss": 0.1319747269153595,
+      "eval_runtime": 688.5799,
+      "eval_samples_per_second": 5.289,
+      "eval_steps_per_second": 0.662,
+      "eval_wer": 0.1540246959428094,
+      "step": 5000
+    },
+    {
+      "epoch": 2.5,
+      "eval_loss": 0.13689081370830536,
+      "eval_runtime": 691.8314,
+      "eval_samples_per_second": 5.264,
+      "eval_steps_per_second": 0.659,
+      "eval_wer": 0.15690279454089684,
+      "step": 5200
+    },
+    {
+      "epoch": 2.6,
+      "eval_loss": 0.13139554858207703,
+      "eval_runtime": 689.159,
+      "eval_samples_per_second": 5.285,
+      "eval_steps_per_second": 0.662,
+      "eval_wer": 0.1516572277411568,
+      "step": 5400
+    },
+    {
+      "epoch": 2.65,
+      "learning_rate": 0.0003431563565513882,
+      "loss": 0.2154,
+      "step": 5500
+    },
+    {
+      "epoch": 2.69,
+      "eval_loss": 0.1303856372833252,
+      "eval_runtime": 694.7157,
+      "eval_samples_per_second": 5.242,
+      "eval_steps_per_second": 0.656,
+      "eval_wer": 0.15063596694828707,
+      "step": 5600
+    },
+    {
+      "epoch": 2.79,
+      "eval_loss": 0.13201411068439484,
+      "eval_runtime": 691.5101,
+      "eval_samples_per_second": 5.267,
+      "eval_steps_per_second": 0.659,
+      "eval_wer": 0.15072880883854795,
+      "step": 5800
+    },
+    {
+      "epoch": 2.89,
+      "learning_rate": 0.0002822698490014613,
+      "loss": 0.2123,
+      "step": 6000
+    },
+    {
+      "epoch": 2.89,
+      "eval_loss": 0.13187964260578156,
+      "eval_runtime": 687.2712,
+      "eval_samples_per_second": 5.299,
+      "eval_steps_per_second": 0.663,
+      "eval_wer": 0.1523999628632439,
+      "step": 6000
+    },
+    {
+      "epoch": 2.98,
+      "eval_loss": 0.12917861342430115,
+      "eval_runtime": 691.2948,
+      "eval_samples_per_second": 5.268,
+      "eval_steps_per_second": 0.66,
+      "eval_wer": 0.1523999628632439,
+      "step": 6200
+    },
+    {
+      "epoch": 3.08,
+      "eval_loss": 0.12825024127960205,
+      "eval_runtime": 689.7813,
+      "eval_samples_per_second": 5.28,
+      "eval_steps_per_second": 0.661,
+      "eval_wer": 0.1488255500881998,
+      "step": 6400
+    },
+    {
+      "epoch": 3.13,
+      "learning_rate": 0.00022138334145153436,
+      "loss": 0.2109,
+      "step": 6500
+    },
+    {
+      "epoch": 3.18,
+      "eval_loss": 0.1257564276456833,
+      "eval_runtime": 687.694,
+      "eval_samples_per_second": 5.296,
+      "eval_steps_per_second": 0.663,
+      "eval_wer": 0.14919691764924334,
+      "step": 6600
+    },
+    {
+      "epoch": 3.27,
+      "eval_loss": 0.12906372547149658,
+      "eval_runtime": 687.8093,
+      "eval_samples_per_second": 5.295,
+      "eval_steps_per_second": 0.663,
+      "eval_wer": 0.1488255500881998,
+      "step": 6800
+    },
+    {
+      "epoch": 3.37,
+      "learning_rate": 0.0001604968339016074,
+      "loss": 0.2103,
+      "step": 7000
+    },
+    {
+      "epoch": 3.37,
+      "eval_loss": 0.12778830528259277,
+      "eval_runtime": 694.3257,
+      "eval_samples_per_second": 5.245,
+      "eval_steps_per_second": 0.657,
+      "eval_wer": 0.14840776158202582,
+      "step": 7000
+    },
+    {
+      "epoch": 3.46,
+      "eval_loss": 0.12501177191734314,
+      "eval_runtime": 679.1124,
+      "eval_samples_per_second": 5.363,
+      "eval_steps_per_second": 0.671,
+      "eval_wer": 0.14780428929533004,
+      "step": 7200
+    },
+    {
+      "epoch": 3.56,
+      "eval_loss": 0.12769711017608643,
+      "eval_runtime": 683.2755,
+      "eval_samples_per_second": 5.33,
+      "eval_steps_per_second": 0.667,
+      "eval_wer": 0.14822207780150404,
+      "step": 7400
+    },
+    {
+      "epoch": 3.61,
+      "learning_rate": 9.961032635168047e-05,
+      "loss": 0.1986,
+      "step": 7500
+    },
+    {
+      "epoch": 3.66,
+      "eval_loss": 0.1256353259086609,
+      "eval_runtime": 680.6384,
+      "eval_samples_per_second": 5.351,
+      "eval_steps_per_second": 0.67,
+      "eval_wer": 0.14757218456967783,
+      "step": 7600
+    },
+    {
+      "epoch": 3.75,
+      "eval_loss": 0.12579868733882904,
+      "eval_runtime": 683.2757,
+      "eval_samples_per_second": 5.33,
+      "eval_steps_per_second": 0.667,
+      "eval_wer": 0.14682944944759074,
+      "step": 7800
+    },
+    {
+      "epoch": 3.85,
+      "learning_rate": 3.884559181685338e-05,
+      "loss": 0.1954,
+      "step": 8000
+    },
+    {
+      "epoch": 3.85,
+      "eval_loss": 0.12557055056095123,
+      "eval_runtime": 690.701,
+      "eval_samples_per_second": 5.273,
+      "eval_steps_per_second": 0.66,
+      "eval_wer": 0.14645808188654721,
+      "step": 8000
+    },
+    {
+      "epoch": 3.95,
+      "eval_loss": 0.12530682981014252,
+      "eval_runtime": 692.3328,
+      "eval_samples_per_second": 5.26,
+      "eval_steps_per_second": 0.659,
+      "eval_wer": 0.1455760839290688,
+      "step": 8200
+    },
+    {
+      "epoch": 4.0,
+      "step": 8312,
+      "total_flos": 1.5580571693960135e+19,
+      "train_loss": 0.7944976037459608,
+      "train_runtime": 46739.0877,
+      "train_samples_per_second": 0.711,
+      "train_steps_per_second": 0.178
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 8312,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 4,
+  "save_steps": 200,
+  "total_flos": 1.5580571693960135e+19,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}