Upload 10 files

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:82a05a7df138ff182263e2e6f0ba1f5e82f9b1a190e63085de22c42fd8931237
+oid sha256:d6b0967b5ab414903ec82cb4eb256dc82f207ea195239a335fbb7c0f1e434f7c
 size 377851056

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7f05c926aab3b5559ee01ae30be37a775d29c2720683d1d6df7d301632c7e60a
+oid sha256:8e5c4216be93855fbc013153efec4d3f85293e3472110511180806826aaf4fe2
 size 722217338

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c1a522037a4b89dbbfead54522650100d82562cbd01c086bc5e3b1c8b42ddc55
+oid sha256:ba2d5edcb13e7e96adbb0ca65cf4d6241e0e76c2d3114a8ecf24fc189d00c1bb
 size 377899102

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3b9e3258117db71c9cd1d1b5432f9dc26b0468be84ff12e8519902ace3644fea
+oid sha256:da279824d75da67ee43e4d57e1cc5b4735e62b87cb30e4ce9cc7d11a432023a9
 size 1064

trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.06431299448013306,
-  "best_model_checkpoint": "./wav2vec2-base-demo/checkpoint-3500",
+  "best_metric": 0.051292333751916885,
+  "best_model_checkpoint": "./wav2vec2-base-demo/checkpoint-4500",
   "epoch": 10.0,
   "eval_steps": 500,
   "global_step": 4810,
@@ -10,182 +10,182 @@
   "log_history": [
     {
       "epoch": 1.0395010395010396,
-      "grad_norm": 2.656402349472046,
-      "learning_rate": 5e-05,
-      "loss": 0.163,
+      "grad_norm": 1.8444721698760986,
+      "learning_rate": 4.99e-05,
+      "loss": 0.1188,
       "step": 500
     },
     {
       "epoch": 1.0395010395010396,
-      "eval_cer": 0.042475895792468986,
-      "eval_loss": 0.0863703265786171,
-      "eval_mer": 0.11671012356875637,
-      "eval_runtime": 37.5343,
-      "eval_samples_per_second": 40.949,
-      "eval_steps_per_second": 5.142,
-      "eval_wer": 0.11718171987934665,
-      "eval_wil": 0.20816339596717104,
-      "eval_wip": 0.791836604032829,
+      "eval_cer": 0.035194462663447716,
+      "eval_loss": 0.06642936915159225,
+      "eval_mer": 0.10481941437848924,
+      "eval_runtime": 34.4969,
+      "eval_samples_per_second": 44.555,
+      "eval_steps_per_second": 5.595,
+      "eval_wer": 0.10539580707984877,
+      "eval_wil": 0.1892106514423464,
+      "eval_wip": 0.8107893485576536,
       "step": 500
     },
     {
       "epoch": 2.079002079002079,
-      "grad_norm": 0.8406820297241211,
-      "learning_rate": 4.4199535962877034e-05,
-      "loss": 0.1635,
+      "grad_norm": 1.2975250482559204,
+      "learning_rate": 4.421113689095128e-05,
+      "loss": 0.1151,
       "step": 1000
     },
     {
       "epoch": 2.079002079002079,
-      "eval_cer": 0.04208912340801724,
-      "eval_loss": 0.07893521338701248,
-      "eval_mer": 0.11592479329482387,
-      "eval_runtime": 37.2156,
-      "eval_samples_per_second": 41.3,
-      "eval_steps_per_second": 5.186,
-      "eval_wer": 0.11649877639292015,
-      "eval_wil": 0.20577031582271066,
-      "eval_wip": 0.7942296841772893,
+      "eval_cer": 0.034538578545611855,
+      "eval_loss": 0.062206532806158066,
+      "eval_mer": 0.10494778291388461,
+      "eval_runtime": 34.1611,
+      "eval_samples_per_second": 44.993,
+      "eval_steps_per_second": 5.65,
+      "eval_wer": 0.10533852674991408,
+      "eval_wil": 0.19021462347071394,
+      "eval_wip": 0.8097853765292861,
       "step": 1000
     },
     {
       "epoch": 3.1185031185031185,
-      "grad_norm": 0.7278923392295837,
-      "learning_rate": 3.839907192575406e-05,
-      "loss": 0.172,
+      "grad_norm": 0.5322809815406799,
+      "learning_rate": 3.8410672853828304e-05,
+      "loss": 0.1235,
       "step": 1500
     },
     {
       "epoch": 3.1185031185031185,
-      "eval_cer": 0.04258640218802663,
-      "eval_loss": 0.09011566638946533,
-      "eval_mer": 0.11518887129608686,
-      "eval_runtime": 37.2272,
-      "eval_samples_per_second": 41.287,
-      "eval_steps_per_second": 5.184,
-      "eval_wer": 0.11592965682089806,
-      "eval_wil": 0.20315250387257455,
-      "eval_wip": 0.7968474961274254,
+      "eval_cer": 0.03377105457793159,
+      "eval_loss": 0.062230244278907776,
+      "eval_mer": 0.10482995141468991,
+      "eval_runtime": 34.2894,
+      "eval_samples_per_second": 44.824,
+      "eval_steps_per_second": 5.629,
+      "eval_wer": 0.10505212510024058,
+      "eval_wil": 0.19100713135160063,
+      "eval_wip": 0.8089928686483994,
       "step": 1500
     },
     {
       "epoch": 4.158004158004158,
-      "grad_norm": 1.2081284523010254,
-      "learning_rate": 3.261020881670534e-05,
-      "loss": 0.148,
+      "grad_norm": 1.327663779258728,
+      "learning_rate": 3.262180974477959e-05,
+      "loss": 0.1068,
       "step": 2000
     },
     {
       "epoch": 4.158004158004158,
-      "eval_cer": 0.04066635356521259,
-      "eval_loss": 0.07975054532289505,
-      "eval_mer": 0.11372148450800136,
-      "eval_runtime": 37.1346,
-      "eval_samples_per_second": 41.39,
-      "eval_steps_per_second": 5.197,
-      "eval_wer": 0.1140515622332252,
-      "eval_wil": 0.2032785755642632,
-      "eval_wip": 0.7967214244357368,
+      "eval_cer": 0.03463626341422571,
+      "eval_loss": 0.06272279471158981,
+      "eval_mer": 0.10423917384606607,
+      "eval_runtime": 34.2637,
+      "eval_samples_per_second": 44.858,
+      "eval_steps_per_second": 5.633,
+      "eval_wer": 0.10465116279069768,
+      "eval_wil": 0.18912080601859016,
+      "eval_wip": 0.8108791939814098,
       "step": 2000
     },
     {
       "epoch": 5.197505197505198,
-      "grad_norm": 1.7628921270370483,
-      "learning_rate": 2.6809744779582367e-05,
-      "loss": 0.1526,
+      "grad_norm": 1.7840451002120972,
+      "learning_rate": 2.6821345707656613e-05,
+      "loss": 0.1123,
       "step": 2500
     },
     {
       "epoch": 5.197505197505198,
-      "eval_cer": 0.04003094179075614,
-      "eval_loss": 0.07782892137765884,
-      "eval_mer": 0.11206896551724138,
-      "eval_runtime": 37.1954,
-      "eval_samples_per_second": 41.322,
-      "eval_steps_per_second": 5.189,
-      "eval_wer": 0.11245802743156337,
-      "eval_wil": 0.2010933536308993,
-      "eval_wip": 0.7989066463691007,
+      "eval_cer": 0.033366360122245635,
+      "eval_loss": 0.06801871210336685,
+      "eval_mer": 0.10333599908602764,
+      "eval_runtime": 34.1023,
+      "eval_samples_per_second": 45.07,
+      "eval_steps_per_second": 5.659,
+      "eval_wer": 0.10362011685187307,
+      "eval_wil": 0.18807839647659041,
+      "eval_wip": 0.8119216035234096,
       "step": 2500
     },
     {
       "epoch": 6.237006237006237,
-      "grad_norm": 1.1964457035064697,
-      "learning_rate": 2.1020881670533645e-05,
-      "loss": 0.1412,
+      "grad_norm": 0.5479459762573242,
+      "learning_rate": 2.1032482598607887e-05,
+      "loss": 0.1066,
       "step": 3000
     },
     {
       "epoch": 6.237006237006237,
-      "eval_cer": 0.03990662209575379,
-      "eval_loss": 0.0688575804233551,
-      "eval_mer": 0.11312191524366029,
-      "eval_runtime": 37.255,
-      "eval_samples_per_second": 41.256,
-      "eval_steps_per_second": 5.181,
-      "eval_wer": 0.11348244266120312,
-      "eval_wil": 0.20307383068892626,
-      "eval_wip": 0.7969261693110737,
+      "eval_cer": 0.03377105457793159,
+      "eval_loss": 0.0692114531993866,
+      "eval_mer": 0.10309631065746706,
+      "eval_runtime": 34.3278,
+      "eval_samples_per_second": 44.774,
+      "eval_steps_per_second": 5.622,
+      "eval_wer": 0.10356283652193837,
+      "eval_wil": 0.18640279212307265,
+      "eval_wip": 0.8135972078769274,
       "step": 3000
     },
     {
       "epoch": 7.276507276507276,
-      "grad_norm": 1.3279238939285278,
-      "learning_rate": 1.523201856148492e-05,
-      "loss": 0.1277,
+      "grad_norm": 2.529324769973755,
+      "learning_rate": 1.5243619489559166e-05,
+      "loss": 0.0985,
       "step": 3500
     },
     {
       "epoch": 7.276507276507276,
-      "eval_cer": 0.03918833052462911,
-      "eval_loss": 0.06431299448013306,
-      "eval_mer": 0.1120195310282178,
-      "eval_runtime": 37.2025,
-      "eval_samples_per_second": 41.314,
-      "eval_steps_per_second": 5.188,
-      "eval_wer": 0.11228729155995675,
-      "eval_wil": 0.20144489838528856,
-      "eval_wip": 0.7985551016147114,
+      "eval_cer": 0.032780250910562525,
+      "eval_loss": 0.0544947050511837,
+      "eval_mer": 0.1027412906910337,
+      "eval_runtime": 34.2034,
+      "eval_samples_per_second": 44.937,
+      "eval_steps_per_second": 5.643,
+      "eval_wer": 0.10304731355252607,
+      "eval_wil": 0.1872840097439975,
+      "eval_wip": 0.8127159902560025,
       "step": 3500
     },
     {
       "epoch": 8.316008316008316,
-      "grad_norm": 1.6859196424484253,
-      "learning_rate": 9.443155452436194e-06,
-      "loss": 0.1376,
+      "grad_norm": 2.9378857612609863,
+      "learning_rate": 9.454756380510442e-06,
+      "loss": 0.1145,
       "step": 4000
     },
     {
       "epoch": 8.316008316008316,
-      "eval_cer": 0.039409343315744395,
-      "eval_loss": 0.06848356872797012,
-      "eval_mer": 0.11210711448995801,
-      "eval_runtime": 37.256,
-      "eval_samples_per_second": 41.255,
-      "eval_steps_per_second": 5.18,
-      "eval_wer": 0.11245802743156337,
-      "eval_wil": 0.20074433269163805,
-      "eval_wip": 0.799255667308362,
+      "eval_cer": 0.03286398079794583,
+      "eval_loss": 0.05142713338136673,
+      "eval_mer": 0.10264465642314503,
+      "eval_runtime": 34.3124,
+      "eval_samples_per_second": 44.794,
+      "eval_steps_per_second": 5.625,
+      "eval_wer": 0.10293275289265666,
+      "eval_wil": 0.18734075205582912,
+      "eval_wip": 0.8126592479441709,
       "step": 4000
     },
     {
       "epoch": 9.355509355509355,
-      "grad_norm": 1.5659629106521606,
-      "learning_rate": 3.642691415313225e-06,
-      "loss": 0.131,
+      "grad_norm": 1.8327735662460327,
+      "learning_rate": 3.654292343387471e-06,
+      "loss": 0.1128,
       "step": 4500
     },
     {
       "epoch": 9.355509355509355,
-      "eval_cer": 0.038787744840732656,
-      "eval_loss": 0.06696277111768723,
-      "eval_mer": 0.11123091765507065,
-      "eval_runtime": 37.5745,
-      "eval_samples_per_second": 40.905,
-      "eval_steps_per_second": 5.136,
-      "eval_wer": 0.11154743611632804,
-      "eval_wil": 0.19943676811944389,
-      "eval_wip": 0.8005632318805561,
+      "eval_cer": 0.03269652102317922,
+      "eval_loss": 0.051292333751916885,
+      "eval_mer": 0.10197669104204753,
+      "eval_runtime": 34.3592,
+      "eval_samples_per_second": 44.733,
+      "eval_steps_per_second": 5.617,
+      "eval_wer": 0.10224538893344026,
+      "eval_wil": 0.1861754557735349,
+      "eval_wip": 0.8138245442264651,
       "step": 4500
     }
   ],

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b3091a22b40a081428e5812146fe9612f2921413ec1d4139f6c30058ff99d057
+oid sha256:adb2dde227c901afcc4b8ad055a16de87aa03de53761c737e130e66fa8b7dd9a
 size 5176