End of training

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1e8848c4dbe8d243b8bcf33ca12e6a97243231e706d3728ac5cd04aefa7ba673
+oid sha256:348ab06f94fbbb9a4199388bb3e4280f3c923b86ec82c2428d19d8ce781444bd
 size 379933779

trainer_state.json

@@ -1,325 +1,250 @@
 {
-  "best_metric": 0.9141282439231873,
-  "best_model_checkpoint": "kurosekurose/wav2vec2-base-EMOPIA/checkpoint-410",
-  "epoch": 19.984,
-  "global_step": 820,
+  "best_metric": 0.7735137939453125,
+  "best_model_checkpoint": "kurosekurose/wav2vec2-base-EMOPIA/checkpoint-348",
+  "epoch": 14.994285714285715,
+  "global_step": 870,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.98,
-      "learning_rate": 1.3666666666666668e-06,
-      "loss": 1.4068,
-      "step": 41
+      "epoch": 0.99,
+      "learning_rate": 1.9333333333333336e-06,
+      "loss": 1.3867,
+      "step": 58
     },
     {
-      "epoch": 0.98,
-      "eval_accuracy": 0.30000001192092896,
-      "eval_loss": 1.370680570602417,
-      "eval_runtime": 40.3938,
-      "eval_samples_per_second": 1.238,
-      "eval_steps_per_second": 0.322,
-      "step": 41
+      "epoch": 0.99,
+      "eval_accuracy": 0.3285714387893677,
+      "eval_loss": 1.3517659902572632,
+      "eval_runtime": 46.5999,
+      "eval_samples_per_second": 1.502,
+      "eval_steps_per_second": 0.386,
+      "step": 58
     },
     {
-      "epoch": 1.98,
-      "learning_rate": 2.7333333333333336e-06,
-      "loss": 1.3568,
-      "step": 82
+      "epoch": 1.99,
+      "learning_rate": 3.866666666666667e-06,
+      "loss": 1.2842,
+      "step": 116
     },
     {
-      "epoch": 1.98,
-      "eval_accuracy": 0.3199999928474426,
-      "eval_loss": 1.3092886209487915,
-      "eval_runtime": 40.4466,
-      "eval_samples_per_second": 1.236,
-      "eval_steps_per_second": 0.321,
-      "step": 82
+      "epoch": 1.99,
+      "eval_accuracy": 0.4571428596973419,
+      "eval_loss": 1.18719482421875,
+      "eval_runtime": 46.74,
+      "eval_samples_per_second": 1.498,
+      "eval_steps_per_second": 0.385,
+      "step": 116
     },
     {
-      "epoch": 2.98,
-      "learning_rate": 4.1e-06,
-      "loss": 1.2503,
-      "step": 123
+      "epoch": 2.99,
+      "learning_rate": 5.8e-06,
+      "loss": 1.0725,
+      "step": 174
     },
     {
-      "epoch": 2.98,
-      "eval_accuracy": 0.46000000834465027,
-      "eval_loss": 1.178250789642334,
-      "eval_runtime": 40.4213,
-      "eval_samples_per_second": 1.237,
-      "eval_steps_per_second": 0.322,
-      "step": 123
+      "epoch": 2.99,
+      "eval_accuracy": 0.5571428537368774,
+      "eval_loss": 1.0476346015930176,
+      "eval_runtime": 46.9576,
+      "eval_samples_per_second": 1.491,
+      "eval_steps_per_second": 0.383,
+      "step": 174
     },
     {
-      "epoch": 3.98,
-      "learning_rate": 5.466666666666667e-06,
-      "loss": 1.1088,
-      "step": 164
+      "epoch": 3.99,
+      "learning_rate": 7.733333333333334e-06,
+      "loss": 0.9343,
+      "step": 232
     },
     {
-      "epoch": 3.98,
-      "eval_accuracy": 0.41999998688697815,
-      "eval_loss": 1.16064453125,
-      "eval_runtime": 40.5701,
-      "eval_samples_per_second": 1.232,
-      "eval_steps_per_second": 0.32,
-      "step": 164
+      "epoch": 3.99,
+      "eval_accuracy": 0.5714285969734192,
+      "eval_loss": 0.9676682949066162,
+      "eval_runtime": 46.68,
+      "eval_samples_per_second": 1.5,
+      "eval_steps_per_second": 0.386,
+      "step": 232
     },
     {
-      "epoch": 4.98,
-      "learning_rate": 6.833333333333334e-06,
-      "loss": 0.9996,
-      "step": 205
+      "epoch": 4.99,
+      "learning_rate": 9.666666666666667e-06,
+      "loss": 0.8053,
+      "step": 290
     },
     {
-      "epoch": 4.98,
-      "eval_accuracy": 0.5400000214576721,
-      "eval_loss": 1.027118444442749,
-      "eval_runtime": 40.2938,
-      "eval_samples_per_second": 1.241,
-      "eval_steps_per_second": 0.323,
-      "step": 205
+      "epoch": 4.99,
+      "eval_accuracy": 0.6142857074737549,
+      "eval_loss": 0.9524909257888794,
+      "eval_runtime": 46.7102,
+      "eval_samples_per_second": 1.499,
+      "eval_steps_per_second": 0.385,
+      "step": 290
     },
     {
-      "epoch": 5.98,
-      "learning_rate": 8.2e-06,
-      "loss": 0.924,
-      "step": 246
+      "epoch": 5.99,
+      "learning_rate": 9.157894736842105e-06,
+      "loss": 0.7895,
+      "step": 348
     },
     {
-      "epoch": 5.98,
-      "eval_accuracy": 0.5799999833106995,
-      "eval_loss": 1.009319543838501,
-      "eval_runtime": 40.7551,
-      "eval_samples_per_second": 1.227,
-      "eval_steps_per_second": 0.319,
-      "step": 246
+      "epoch": 5.99,
+      "eval_accuracy": 0.6857143044471741,
+      "eval_loss": 0.7735137939453125,
+      "eval_runtime": 46.9896,
+      "eval_samples_per_second": 1.49,
+      "eval_steps_per_second": 0.383,
+      "step": 348
     },
     {
-      "epoch": 6.98,
-      "learning_rate": 9.566666666666668e-06,
-      "loss": 0.8449,
-      "step": 287
+      "epoch": 6.99,
+      "learning_rate": 8.140350877192983e-06,
+      "loss": 0.6867,
+      "step": 406
     },
     {
-      "epoch": 6.98,
-      "eval_accuracy": 0.5799999833106995,
-      "eval_loss": 0.961348295211792,
-      "eval_runtime": 40.5666,
-      "eval_samples_per_second": 1.233,
-      "eval_steps_per_second": 0.32,
-      "step": 287
+      "epoch": 6.99,
+      "eval_accuracy": 0.6428571343421936,
+      "eval_loss": 0.8555552363395691,
+      "eval_runtime": 46.718,
+      "eval_samples_per_second": 1.498,
+      "eval_steps_per_second": 0.385,
+      "step": 406
     },
     {
-      "epoch": 7.98,
-      "learning_rate": 9.461538461538463e-06,
-      "loss": 0.7559,
-      "step": 328
+      "epoch": 7.99,
+      "learning_rate": 7.140350877192983e-06,
+      "loss": 0.6218,
+      "step": 464
     },
     {
-      "epoch": 7.98,
-      "eval_accuracy": 0.6000000238418579,
-      "eval_loss": 0.937028169631958,
-      "eval_runtime": 40.5792,
-      "eval_samples_per_second": 1.232,
-      "eval_steps_per_second": 0.32,
-      "step": 328
+      "epoch": 7.99,
+      "eval_accuracy": 0.6714285612106323,
+      "eval_loss": 0.8454030156135559,
+      "eval_runtime": 46.4466,
+      "eval_samples_per_second": 1.507,
+      "eval_steps_per_second": 0.388,
+      "step": 464
     },
     {
-      "epoch": 8.98,
-      "learning_rate": 8.673076923076924e-06,
-      "loss": 0.7744,
-      "step": 369
+      "epoch": 8.99,
+      "learning_rate": 6.140350877192983e-06,
+      "loss": 0.558,
+      "step": 522
     },
     {
-      "epoch": 8.98,
-      "eval_accuracy": 0.5600000023841858,
-      "eval_loss": 1.0094345808029175,
-      "eval_runtime": 40.4365,
-      "eval_samples_per_second": 1.237,
-      "eval_steps_per_second": 0.321,
-      "step": 369
+      "epoch": 8.99,
+      "eval_accuracy": 0.6571428775787354,
+      "eval_loss": 0.8404667973518372,
+      "eval_runtime": 46.7611,
+      "eval_samples_per_second": 1.497,
+      "eval_steps_per_second": 0.385,
+      "step": 522
     },
     {
-      "epoch": 9.98,
-      "learning_rate": 7.884615384615384e-06,
-      "loss": 0.6745,
-      "step": 410
+      "epoch": 9.99,
+      "learning_rate": 5.12280701754386e-06,
+      "loss": 0.5033,
+      "step": 580
     },
     {
-      "epoch": 9.98,
-      "eval_accuracy": 0.6399999856948853,
-      "eval_loss": 0.9141282439231873,
-      "eval_runtime": 40.471,
-      "eval_samples_per_second": 1.235,
-      "eval_steps_per_second": 0.321,
-      "step": 410
+      "epoch": 9.99,
+      "eval_accuracy": 0.6285714507102966,
+      "eval_loss": 1.0190438032150269,
+      "eval_runtime": 47.1353,
+      "eval_samples_per_second": 1.485,
+      "eval_steps_per_second": 0.382,
+      "step": 580
     },
     {
-      "epoch": 10.98,
-      "learning_rate": 7.096153846153847e-06,
-      "loss": 0.6245,
-      "step": 451
+      "epoch": 10.99,
+      "learning_rate": 4.105263157894737e-06,
+      "loss": 0.4403,
+      "step": 638
     },
     {
-      "epoch": 10.98,
-      "eval_accuracy": 0.5,
-      "eval_loss": 1.1951998472213745,
-      "eval_runtime": 40.6687,
-      "eval_samples_per_second": 1.229,
-      "eval_steps_per_second": 0.32,
-      "step": 451
+      "epoch": 10.99,
+      "eval_accuracy": 0.699999988079071,
+      "eval_loss": 0.82489013671875,
+      "eval_runtime": 46.7633,
+      "eval_samples_per_second": 1.497,
+      "eval_steps_per_second": 0.385,
+      "step": 638
     },
     {
-      "epoch": 11.98,
-      "learning_rate": 6.307692307692308e-06,
-      "loss": 0.5174,
-      "step": 492
+      "epoch": 11.99,
+      "learning_rate": 3.0877192982456146e-06,
+      "loss": 0.3995,
+      "step": 696
     },
     {
-      "epoch": 11.98,
-      "eval_accuracy": 0.6600000262260437,
-      "eval_loss": 0.9249220490455627,
-      "eval_runtime": 40.488,
-      "eval_samples_per_second": 1.235,
-      "eval_steps_per_second": 0.321,
-      "step": 492
+      "epoch": 11.99,
+      "eval_accuracy": 0.7142857313156128,
+      "eval_loss": 0.8996997475624084,
+      "eval_runtime": 46.7333,
+      "eval_samples_per_second": 1.498,
+      "eval_steps_per_second": 0.385,
+      "step": 696
     },
     {
-      "epoch": 12.98,
-      "learning_rate": 5.5192307692307705e-06,
-      "loss": 0.5091,
-      "step": 533
+      "epoch": 12.99,
+      "learning_rate": 2.0701754385964913e-06,
+      "loss": 0.3534,
+      "step": 754
     },
     {
-      "epoch": 12.98,
-      "eval_accuracy": 0.7200000286102295,
-      "eval_loss": 0.9292579293251038,
-      "eval_runtime": 40.6346,
-      "eval_samples_per_second": 1.23,
-      "eval_steps_per_second": 0.32,
-      "step": 533
+      "epoch": 12.99,
+      "eval_accuracy": 0.699999988079071,
+      "eval_loss": 0.9177280068397522,
+      "eval_runtime": 46.7502,
+      "eval_samples_per_second": 1.497,
+      "eval_steps_per_second": 0.385,
+      "step": 754
     },
     {
-      "epoch": 13.98,
-      "learning_rate": 4.730769230769231e-06,
-      "loss": 0.391,
-      "step": 574
+      "epoch": 13.99,
+      "learning_rate": 1.0526315789473685e-06,
+      "loss": 0.3023,
+      "step": 812
     },
     {
-      "epoch": 13.98,
-      "eval_accuracy": 0.6600000262260437,
-      "eval_loss": 0.9355679750442505,
-      "eval_runtime": 40.4106,
-      "eval_samples_per_second": 1.237,
-      "eval_steps_per_second": 0.322,
-      "step": 574
+      "epoch": 13.99,
+      "eval_accuracy": 0.6571428775787354,
+      "eval_loss": 0.9544177651405334,
+      "eval_runtime": 46.7592,
+      "eval_samples_per_second": 1.497,
+      "eval_steps_per_second": 0.385,
+      "step": 812
     },
     {
-      "epoch": 14.98,
-      "learning_rate": 3.961538461538462e-06,
-      "loss": 0.3689,
-      "step": 615
+      "epoch": 14.99,
+      "learning_rate": 3.508771929824562e-08,
+      "loss": 0.2752,
+      "step": 870
     },
     {
-      "epoch": 14.98,
-      "eval_accuracy": 0.6200000047683716,
-      "eval_loss": 1.0244715213775635,
-      "eval_runtime": 40.302,
-      "eval_samples_per_second": 1.241,
-      "eval_steps_per_second": 0.323,
-      "step": 615
+      "epoch": 14.99,
+      "eval_accuracy": 0.699999988079071,
+      "eval_loss": 0.9346058964729309,
+      "eval_runtime": 46.4397,
+      "eval_samples_per_second": 1.507,
+      "eval_steps_per_second": 0.388,
+      "step": 870
     },
     {
-      "epoch": 15.98,
-      "learning_rate": 3.1730769230769233e-06,
-      "loss": 0.2816,
-      "step": 656
-    },
-    {
-      "epoch": 15.98,
-      "eval_accuracy": 0.6600000262260437,
-      "eval_loss": 1.0420873165130615,
-      "eval_runtime": 40.6297,
-      "eval_samples_per_second": 1.231,
-      "eval_steps_per_second": 0.32,
-      "step": 656
-    },
-    {
-      "epoch": 16.98,
-      "learning_rate": 2.384615384615385e-06,
-      "loss": 0.3003,
-      "step": 697
-    },
-    {
-      "epoch": 16.98,
-      "eval_accuracy": 0.6399999856948853,
-      "eval_loss": 1.042699933052063,
-      "eval_runtime": 40.2488,
-      "eval_samples_per_second": 1.242,
-      "eval_steps_per_second": 0.323,
-      "step": 697
-    },
-    {
-      "epoch": 17.98,
-      "learning_rate": 1.596153846153846e-06,
-      "loss": 0.2495,
-      "step": 738
-    },
-    {
-      "epoch": 17.98,
-      "eval_accuracy": 0.6600000262260437,
-      "eval_loss": 1.095711350440979,
-      "eval_runtime": 40.5759,
-      "eval_samples_per_second": 1.232,
-      "eval_steps_per_second": 0.32,
-      "step": 738
-    },
-    {
-      "epoch": 18.98,
-      "learning_rate": 8.076923076923078e-07,
-      "loss": 0.2269,
-      "step": 779
-    },
-    {
-      "epoch": 18.98,
-      "eval_accuracy": 0.6600000262260437,
-      "eval_loss": 1.05389404296875,
-      "eval_runtime": 40.5913,
-      "eval_samples_per_second": 1.232,
-      "eval_steps_per_second": 0.32,
-      "step": 779
-    },
-    {
-      "epoch": 19.98,
-      "learning_rate": 1.9230769230769234e-08,
-      "loss": 0.2055,
-      "step": 820
-    },
-    {
-      "epoch": 19.98,
-      "eval_accuracy": 0.6800000071525574,
-      "eval_loss": 1.062558650970459,
-      "eval_runtime": 40.7382,
-      "eval_samples_per_second": 1.227,
-      "eval_steps_per_second": 0.319,
-      "step": 820
-    },
-    {
-      "epoch": 19.98,
-      "step": 820,
-      "total_flos": 5.721848459887288e+18,
-      "train_loss": 0.6885371266341791,
-      "train_runtime": 22708.6522,
-      "train_samples_per_second": 0.44,
-      "train_steps_per_second": 0.036
+      "epoch": 14.99,
+      "step": 870,
+      "total_flos": 4.99147314409986e+18,
+      "train_loss": 0.6941862314596944,
+      "train_runtime": 20182.4464,
+      "train_samples_per_second": 0.52,
+      "train_steps_per_second": 0.043
     }
   ],
-  "max_steps": 820,
-  "num_train_epochs": 20,
-  "total_flos": 5.721848459887288e+18,
+  "max_steps": 870,
+  "num_train_epochs": 15,
+  "total_flos": 4.99147314409986e+18,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9ccf48c4a0d78c84d927241619accb6e146ed3470c5f78ef19ec300caa18ccaa
+oid sha256:129989de89a833eb0c8d9e960516ce76d7fddab79482e76bde4c7c1f16fb577e
 size 3055