Upload 10 files

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:60742fd4afd2bfa2b90aab0208a23c41d1b8e00609456f357ca1fcbc8fd4fab7
+oid sha256:8100fede3ebe12028e1a6a943068124b87c44e904129909600f8aa61534da523
 size 1264702096

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f9c4dfbab4dc439b8aa429bc9e96d6cc56150ae914389a765cf958a5895bb734
+oid sha256:8ffba324e21204e3e61ec695badb71f79f68c8cfd8af041202e4e0dbfad6ee65
 size 1264796970

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:874049e61979cf2df2a665c44cd0ef1f2f2cfe53bf0dc13ced6b402c6ef50d57
+oid sha256:ad8e0e131aa62acb514576ec4fc6d35c4209b434e53cc06404b8837a81316972
 size 14244

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1c064ab2f936676a9b58c77a232957298d190a50d5663df18e93d1d5b1a9a231
+oid sha256:b37cc57efe149b952d1d08baf340d5ff65995cd05283b74a9e6abf14b366d6de
 size 1064

trainer_state.json

@@ -1,198 +1,98 @@
 {
-  "best_metric": 0.16613277792930603,
-  "best_model_checkpoint": "./wav2vec2-base-demo/checkpoint-4500",
-  "epoch": 2.0,
+  "best_metric": 0.14490656554698944,
+  "best_model_checkpoint": "./wav2vec2-base-demo/checkpoint-2000",
+  "epoch": 0.986679822397632,
   "eval_steps": 500,
-  "global_step": 4864,
+  "global_step": 2000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.20559210526315788,
-      "grad_norm": 1.0973399877548218,
+      "epoch": 0.246669955599408,
+      "grad_norm": 3.482757329940796,
       "learning_rate": 4.97e-05,
-      "loss": 0.2869,
+      "loss": 0.445,
       "step": 500
     },
     {
-      "epoch": 0.20559210526315788,
-      "eval_cer": 0.03840555601763287,
-      "eval_loss": 0.18634851276874542,
-      "eval_mer": 0.23470839260312945,
-      "eval_runtime": 69.1687,
-      "eval_samples_per_second": 17.58,
-      "eval_steps_per_second": 2.198,
-      "eval_wer": 0.2378892733564014,
-      "eval_wil": 0.39643215927831765,
-      "eval_wip": 0.6035678407216823,
+      "epoch": 0.246669955599408,
+      "eval_cer": 0.03472612578852039,
+      "eval_loss": 0.1506977081298828,
+      "eval_mer": 0.2400241837968561,
+      "eval_runtime": 72.835,
+      "eval_samples_per_second": 16.695,
+      "eval_steps_per_second": 2.087,
+      "eval_wer": 0.24393241167434715,
+      "eval_wil": 0.4027411629459634,
+      "eval_wip": 0.5972588370540366,
       "step": 500
     },
     {
-      "epoch": 0.41118421052631576,
-      "grad_norm": 3.03368878364563,
-      "learning_rate": 4.4305682859761687e-05,
-      "loss": 0.1975,
+      "epoch": 0.493339911198816,
+      "grad_norm": 2.1172962188720703,
+      "learning_rate": 4.5547392940333276e-05,
+      "loss": 0.4275,
       "step": 1000
     },
     {
-      "epoch": 0.41118421052631576,
-      "eval_cer": 0.03877983864260168,
-      "eval_loss": 0.19032993912696838,
-      "eval_mer": 0.2419536984754376,
-      "eval_runtime": 69.0115,
-      "eval_samples_per_second": 17.62,
-      "eval_steps_per_second": 2.203,
-      "eval_wer": 0.24711649365628605,
-      "eval_wil": 0.4058917970631253,
-      "eval_wip": 0.5941082029368747,
+      "epoch": 0.493339911198816,
+      "eval_cer": 0.03416401224158391,
+      "eval_loss": 0.15176725387573242,
+      "eval_mer": 0.23524069028156222,
+      "eval_runtime": 72.5697,
+      "eval_samples_per_second": 16.756,
+      "eval_steps_per_second": 2.095,
+      "eval_wer": 0.23870967741935484,
+      "eval_wil": 0.3973965187752956,
+      "eval_wip": 0.6026034812247044,
       "step": 1000
     },
     {
-      "epoch": 0.6167763157894737,
-      "grad_norm": 1.8283483982086182,
-      "learning_rate": 3.8588450962419795e-05,
-      "loss": 0.2158,
+      "epoch": 0.740009866798224,
+      "grad_norm": 3.9609649181365967,
+      "learning_rate": 4.1067908976885865e-05,
+      "loss": 0.4367,
       "step": 1500
     },
     {
-      "epoch": 0.6167763157894737,
-      "eval_cer": 0.03936205605921983,
-      "eval_loss": 0.1854981929063797,
-      "eval_mer": 0.24507845934379457,
-      "eval_runtime": 69.033,
-      "eval_samples_per_second": 17.615,
-      "eval_steps_per_second": 2.202,
-      "eval_wer": 0.24769319492502884,
-      "eval_wil": 0.4122754264819207,
-      "eval_wip": 0.5877245735180793,
+      "epoch": 0.740009866798224,
+      "eval_cer": 0.032727499843857345,
+      "eval_loss": 0.14839377999305725,
+      "eval_mer": 0.21904472163066627,
+      "eval_runtime": 72.6193,
+      "eval_samples_per_second": 16.745,
+      "eval_steps_per_second": 2.093,
+      "eval_wer": 0.22119815668202766,
+      "eval_wil": 0.3745997578877486,
+      "eval_wip": 0.6254002421122514,
       "step": 1500
     },
     {
-      "epoch": 0.8223684210526315,
-      "grad_norm": 2.217629909515381,
-      "learning_rate": 3.2859761686526126e-05,
-      "loss": 0.2063,
+      "epoch": 0.986679822397632,
+      "grad_norm": 3.834362268447876,
+      "learning_rate": 3.6588425013438454e-05,
+      "loss": 0.4128,
       "step": 2000
     },
     {
-      "epoch": 0.8223684210526315,
-      "eval_cer": 0.03709556683024204,
-      "eval_loss": 0.1818998157978058,
-      "eval_mer": 0.22879908935685828,
-      "eval_runtime": 68.9653,
-      "eval_samples_per_second": 17.632,
-      "eval_steps_per_second": 2.204,
-      "eval_wer": 0.23183391003460208,
-      "eval_wil": 0.38724655848178047,
-      "eval_wip": 0.6127534415182195,
+      "epoch": 0.986679822397632,
+      "eval_cer": 0.032602585722315905,
+      "eval_loss": 0.14490656554698944,
+      "eval_mer": 0.21031746031746032,
+      "eval_runtime": 72.7757,
+      "eval_samples_per_second": 16.709,
+      "eval_steps_per_second": 2.089,
+      "eval_wer": 0.21167434715821812,
+      "eval_wil": 0.3614631097880948,
+      "eval_wip": 0.6385368902119052,
       "step": 2000
-    },
-    {
-      "epoch": 1.0279605263157894,
-      "grad_norm": 1.560766577720642,
-      "learning_rate": 2.7142529789184235e-05,
-      "loss": 0.2183,
-      "step": 2500
-    },
-    {
-      "epoch": 1.0279605263157894,
-      "eval_cer": 0.036617316809448554,
-      "eval_loss": 0.1806473284959793,
-      "eval_mer": 0.22477976697925547,
-      "eval_runtime": 69.4484,
-      "eval_samples_per_second": 17.509,
-      "eval_steps_per_second": 2.189,
-      "eval_wer": 0.22808535178777392,
-      "eval_wil": 0.38140651271785064,
-      "eval_wip": 0.6185934872821494,
-      "step": 2500
-    },
-    {
-      "epoch": 1.2335526315789473,
-      "grad_norm": 0.77958744764328,
-      "learning_rate": 2.141384051329056e-05,
-      "loss": 0.2221,
-      "step": 3000
-    },
-    {
-      "epoch": 1.2335526315789473,
-      "eval_cer": 0.036326208101139484,
-      "eval_loss": 0.17762529850006104,
-      "eval_mer": 0.22323722523551243,
-      "eval_runtime": 68.9232,
-      "eval_samples_per_second": 17.643,
-      "eval_steps_per_second": 2.205,
-      "eval_wer": 0.22549019607843138,
-      "eval_wil": 0.3801096837587259,
-      "eval_wip": 0.6198903162412741,
-      "step": 3000
-    },
-    {
-      "epoch": 1.4391447368421053,
-      "grad_norm": 1.003024935722351,
-      "learning_rate": 1.5685151237396883e-05,
-      "loss": 0.2344,
-      "step": 3500
-    },
-    {
-      "epoch": 1.4391447368421053,
-      "eval_cer": 0.03588954503867587,
-      "eval_loss": 0.17285175621509552,
-      "eval_mer": 0.22013116623895068,
-      "eval_runtime": 69.0254,
-      "eval_samples_per_second": 17.617,
-      "eval_steps_per_second": 2.202,
-      "eval_wer": 0.22260668973471742,
-      "eval_wil": 0.3742594650447425,
-      "eval_wip": 0.6257405349552575,
-      "step": 3500
-    },
-    {
-      "epoch": 1.6447368421052633,
-      "grad_norm": 0.9715732932090759,
-      "learning_rate": 9.956461961503208e-06,
-      "loss": 0.2642,
-      "step": 4000
-    },
-    {
-      "epoch": 1.6447368421052633,
-      "eval_cer": 0.03532812110122266,
-      "eval_loss": 0.16884012520313263,
-      "eval_mer": 0.21776259607173357,
-      "eval_runtime": 69.1041,
-      "eval_samples_per_second": 17.597,
-      "eval_steps_per_second": 2.2,
-      "eval_wer": 0.22058823529411764,
-      "eval_wil": 0.3699418492887351,
-      "eval_wip": 0.6300581507112649,
-      "step": 4000
-    },
-    {
-      "epoch": 1.850328947368421,
-      "grad_norm": 5.113919734954834,
-      "learning_rate": 4.227772685609533e-06,
-      "loss": 0.2684,
-      "step": 4500
-    },
-    {
-      "epoch": 1.850328947368421,
-      "eval_cer": 0.03497463195541878,
-      "eval_loss": 0.16613277792930603,
-      "eval_mer": 0.21094640820980615,
-      "eval_runtime": 69.7332,
-      "eval_samples_per_second": 17.438,
-      "eval_steps_per_second": 2.18,
-      "eval_wer": 0.21337946943483277,
-      "eval_wil": 0.35981112406781113,
-      "eval_wip": 0.6401888759321889,
-      "step": 4500
     }
   ],
   "logging_steps": 500,
-  "max_steps": 4864,
+  "max_steps": 6081,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 2,
+  "num_train_epochs": 3,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -201,13 +101,13 @@
         "should_evaluate": false,
         "should_log": false,
         "should_save": true,
-        "should_training_stop": true
+        "should_training_stop": false
       },
       "attributes": {}
     }
   },
-  "total_flos": 6.660365150995041e+18,
-  "train_batch_size": 10,
+  "total_flos": 3.2957716566380933e+18,
+  "train_batch_size": 12,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4e1cfcc7c7576819400a0be686a2f2e2fb83f17aca5d2cc029e3a9f046772832
+oid sha256:1456fbb80fdcd6a8b153f11eeae0fef4bc2173125f866d139aac039bc0656f69
 size 5176