End of training

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fbd58cd935301cd872d7f66acefe198815b34137c7737f8630f292d3366423a4
+oid sha256:7457c6eea8ce91eb1e913e2bdae191fa9f401dce20c54b7f58078596e6ecc698
 size 379933779

trainer_state.json

@@ -1,325 +1,175 @@
 {
-  "best_metric": 0.8695746660232544,
-  "best_model_checkpoint": "kurosekurose/wav2vec2-base-EMOPIA/checkpoint-522",
-  "epoch": 19.994285714285713,
-  "global_step": 1160,
+  "best_metric": 0.9727444052696228,
+  "best_model_checkpoint": "kurosekurose/wav2vec2-base-EMOPIA/checkpoint-525",
+  "epoch": 10.0,
+  "global_step": 1750,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.99,
-      "learning_rate": 1.1600000000000001e-06,
-      "loss": 1.3846,
-      "step": 58
+      "epoch": 1.0,
+      "learning_rate": 3.5e-06,
+      "loss": 1.3252,
+      "step": 175
     },
     {
-      "epoch": 0.99,
-      "eval_accuracy": 0.41428571939468384,
-      "eval_loss": 1.3525933027267456,
-      "eval_runtime": 48.4658,
-      "eval_samples_per_second": 1.444,
-      "eval_steps_per_second": 0.371,
-      "step": 58
+      "epoch": 1.0,
+      "eval_accuracy": 0.4285714328289032,
+      "eval_loss": 1.2177598476409912,
+      "eval_runtime": 47.4346,
+      "eval_samples_per_second": 1.476,
+      "eval_steps_per_second": 0.379,
+      "step": 175
     },
     {
-      "epoch": 1.99,
-      "learning_rate": 2.3200000000000002e-06,
-      "loss": 1.3182,
-      "step": 116
+      "epoch": 2.0,
+      "learning_rate": 6.98e-06,
+      "loss": 1.039,
+      "step": 350
     },
     {
-      "epoch": 1.99,
-      "eval_accuracy": 0.48571428656578064,
-      "eval_loss": 1.2529839277267456,
-      "eval_runtime": 48.2009,
-      "eval_samples_per_second": 1.452,
-      "eval_steps_per_second": 0.373,
-      "step": 116
+      "epoch": 2.0,
+      "eval_accuracy": 0.5714285969734192,
+      "eval_loss": 1.0262044668197632,
+      "eval_runtime": 47.2654,
+      "eval_samples_per_second": 1.481,
+      "eval_steps_per_second": 0.381,
+      "step": 350
     },
     {
-      "epoch": 2.99,
-      "learning_rate": 3.48e-06,
-      "loss": 1.1606,
-      "step": 174
+      "epoch": 3.0,
+      "learning_rate": 9.808000000000002e-06,
+      "loss": 0.9232,
+      "step": 525
     },
     {
-      "epoch": 2.99,
-      "eval_accuracy": 0.4000000059604645,
-      "eval_loss": 1.0961982011795044,
-      "eval_runtime": 48.4788,
-      "eval_samples_per_second": 1.444,
-      "eval_steps_per_second": 0.371,
-      "step": 174
+      "epoch": 3.0,
+      "eval_accuracy": 0.5714285969734192,
+      "eval_loss": 0.9727444052696228,
+      "eval_runtime": 47.5019,
+      "eval_samples_per_second": 1.474,
+      "eval_steps_per_second": 0.379,
+      "step": 525
     },
     {
-      "epoch": 3.99,
-      "learning_rate": 4.6400000000000005e-06,
-      "loss": 1.0104,
-      "step": 232
+      "epoch": 4.0,
+      "learning_rate": 8.408e-06,
+      "loss": 0.8248,
+      "step": 700
     },
     {
-      "epoch": 3.99,
-      "eval_accuracy": 0.44285714626312256,
-      "eval_loss": 1.0213487148284912,
-      "eval_runtime": 48.0421,
-      "eval_samples_per_second": 1.457,
-      "eval_steps_per_second": 0.375,
-      "step": 232
+      "epoch": 4.0,
+      "eval_accuracy": 0.5142857432365417,
+      "eval_loss": 1.0024999380111694,
+      "eval_runtime": 47.3119,
+      "eval_samples_per_second": 1.48,
+      "eval_steps_per_second": 0.38,
+      "step": 700
     },
     {
-      "epoch": 4.99,
-      "learning_rate": 5.8e-06,
-      "loss": 0.9127,
-      "step": 290
+      "epoch": 5.0,
+      "learning_rate": 7.0080000000000005e-06,
+      "loss": 0.7389,
+      "step": 875
     },
     {
-      "epoch": 4.99,
-      "eval_accuracy": 0.48571428656578064,
-      "eval_loss": 0.9955207109451294,
-      "eval_runtime": 48.3011,
-      "eval_samples_per_second": 1.449,
-      "eval_steps_per_second": 0.373,
-      "step": 290
-    },
-    {
-      "epoch": 5.99,
-      "learning_rate": 6.96e-06,
-      "loss": 0.8464,
-      "step": 348
-    },
-    {
-      "epoch": 5.99,
-      "eval_accuracy": 0.5857142806053162,
-      "eval_loss": 0.9400736689567566,
-      "eval_runtime": 48.3878,
-      "eval_samples_per_second": 1.447,
-      "eval_steps_per_second": 0.372,
-      "step": 348
-    },
-    {
-      "epoch": 6.99,
-      "learning_rate": 8.120000000000002e-06,
-      "loss": 0.787,
-      "step": 406
-    },
-    {
-      "epoch": 6.99,
-      "eval_accuracy": 0.6571428775787354,
-      "eval_loss": 0.9043138027191162,
-      "eval_runtime": 48.4438,
-      "eval_samples_per_second": 1.445,
-      "eval_steps_per_second": 0.372,
-      "step": 406
-    },
-    {
-      "epoch": 7.99,
-      "learning_rate": 9.260000000000001e-06,
-      "loss": 0.7181,
-      "step": 464
-    },
-    {
-      "epoch": 7.99,
-      "eval_accuracy": 0.6000000238418579,
-      "eval_loss": 0.9584859013557434,
-      "eval_runtime": 48.0745,
-      "eval_samples_per_second": 1.456,
-      "eval_steps_per_second": 0.374,
-      "step": 464
-    },
-    {
-      "epoch": 8.99,
-      "learning_rate": 9.681818181818182e-06,
-      "loss": 0.6804,
-      "step": 522
-    },
-    {
-      "epoch": 8.99,
-      "eval_accuracy": 0.6285714507102966,
-      "eval_loss": 0.8695746660232544,
-      "eval_runtime": 48.2058,
-      "eval_samples_per_second": 1.452,
-      "eval_steps_per_second": 0.373,
-      "step": 522
-    },
-    {
-      "epoch": 9.99,
-      "learning_rate": 8.818181818181819e-06,
-      "loss": 0.6065,
-      "step": 580
-    },
-    {
-      "epoch": 9.99,
-      "eval_accuracy": 0.6285714507102966,
-      "eval_loss": 0.9474284052848816,
-      "eval_runtime": 48.413,
-      "eval_samples_per_second": 1.446,
-      "eval_steps_per_second": 0.372,
-      "step": 580
+      "epoch": 5.0,
+      "eval_accuracy": 0.6428571343421936,
+      "eval_loss": 0.9947425127029419,
+      "eval_runtime": 47.6424,
+      "eval_samples_per_second": 1.469,
+      "eval_steps_per_second": 0.378,
+      "step": 875
     },
     {
-      "epoch": 10.99,
-      "learning_rate": 7.93939393939394e-06,
-      "loss": 0.6316,
-      "step": 638
+      "epoch": 6.0,
+      "learning_rate": 5.616e-06,
+      "loss": 0.6253,
+      "step": 1050
     },
     {
-      "epoch": 10.99,
+      "epoch": 6.0,
       "eval_accuracy": 0.6857143044471741,
-      "eval_loss": 0.9475699663162231,
-      "eval_runtime": 47.8485,
-      "eval_samples_per_second": 1.463,
-      "eval_steps_per_second": 0.376,
-      "step": 638
-    },
-    {
-      "epoch": 11.99,
-      "learning_rate": 7.060606060606061e-06,
-      "loss": 0.517,
-      "step": 696
+      "eval_loss": 1.0144152641296387,
+      "eval_runtime": 47.3265,
+      "eval_samples_per_second": 1.479,
+      "eval_steps_per_second": 0.38,
+      "step": 1050
     },
     {
-      "epoch": 11.99,
-      "eval_accuracy": 0.6571428775787354,
-      "eval_loss": 0.9657440185546875,
-      "eval_runtime": 48.3215,
-      "eval_samples_per_second": 1.449,
-      "eval_steps_per_second": 0.373,
-      "step": 696
+      "epoch": 7.0,
+      "learning_rate": 4.216e-06,
+      "loss": 0.5691,
+      "step": 1225
     },
     {
-      "epoch": 12.99,
-      "learning_rate": 6.181818181818182e-06,
-      "loss": 0.4487,
-      "step": 754
-    },
-    {
-      "epoch": 12.99,
-      "eval_accuracy": 0.6428571343421936,
-      "eval_loss": 1.0147874355316162,
-      "eval_runtime": 48.212,
-      "eval_samples_per_second": 1.452,
-      "eval_steps_per_second": 0.373,
-      "step": 754
-    },
-    {
-      "epoch": 13.99,
-      "learning_rate": 5.303030303030303e-06,
-      "loss": 0.4093,
-      "step": 812
-    },
-    {
-      "epoch": 13.99,
+      "epoch": 7.0,
       "eval_accuracy": 0.6714285612106323,
-      "eval_loss": 0.9352009892463684,
-      "eval_runtime": 47.9193,
-      "eval_samples_per_second": 1.461,
-      "eval_steps_per_second": 0.376,
-      "step": 812
-    },
-    {
-      "epoch": 14.99,
-      "learning_rate": 4.424242424242425e-06,
-      "loss": 0.3458,
-      "step": 870
-    },
-    {
-      "epoch": 14.99,
-      "eval_accuracy": 0.6571428775787354,
-      "eval_loss": 1.01514732837677,
-      "eval_runtime": 47.8767,
-      "eval_samples_per_second": 1.462,
-      "eval_steps_per_second": 0.376,
-      "step": 870
-    },
-    {
-      "epoch": 15.99,
-      "learning_rate": 3.5454545454545458e-06,
-      "loss": 0.3082,
-      "step": 928
+      "eval_loss": 1.0645909309387207,
+      "eval_runtime": 47.3515,
+      "eval_samples_per_second": 1.478,
+      "eval_steps_per_second": 0.38,
+      "step": 1225
     },
     {
-      "epoch": 15.99,
-      "eval_accuracy": 0.6571428775787354,
-      "eval_loss": 1.0875554084777832,
-      "eval_runtime": 47.9039,
-      "eval_samples_per_second": 1.461,
-      "eval_steps_per_second": 0.376,
-      "step": 928
+      "epoch": 8.0,
+      "learning_rate": 2.824e-06,
+      "loss": 0.5264,
+      "step": 1400
     },
     {
-      "epoch": 16.99,
-      "learning_rate": 2.666666666666667e-06,
-      "loss": 0.2619,
-      "step": 986
-    },
-    {
-      "epoch": 16.99,
-      "eval_accuracy": 0.699999988079071,
-      "eval_loss": 0.9961318969726562,
-      "eval_runtime": 47.6669,
-      "eval_samples_per_second": 1.469,
-      "eval_steps_per_second": 0.378,
-      "step": 986
-    },
-    {
-      "epoch": 17.99,
-      "learning_rate": 1.787878787878788e-06,
-      "loss": 0.2332,
-      "step": 1044
-    },
-    {
-      "epoch": 17.99,
+      "epoch": 8.0,
       "eval_accuracy": 0.6714285612106323,
-      "eval_loss": 1.15262770652771,
-      "eval_runtime": 48.0091,
-      "eval_samples_per_second": 1.458,
-      "eval_steps_per_second": 0.375,
-      "step": 1044
+      "eval_loss": 1.2786132097244263,
+      "eval_runtime": 47.1262,
+      "eval_samples_per_second": 1.485,
+      "eval_steps_per_second": 0.382,
+      "step": 1400
     },
     {
-      "epoch": 18.99,
-      "learning_rate": 9.090909090909091e-07,
-      "loss": 0.2029,
-      "step": 1102
+      "epoch": 9.0,
+      "learning_rate": 1.424e-06,
+      "loss": 0.4535,
+      "step": 1575
     },
     {
-      "epoch": 18.99,
-      "eval_accuracy": 0.699999988079071,
-      "eval_loss": 1.0439351797103882,
-      "eval_runtime": 47.7114,
-      "eval_samples_per_second": 1.467,
-      "eval_steps_per_second": 0.377,
-      "step": 1102
-    },
-    {
-      "epoch": 19.99,
-      "learning_rate": 3.0303030303030305e-08,
-      "loss": 0.1781,
-      "step": 1160
-    },
-    {
-      "epoch": 19.99,
-      "eval_accuracy": 0.6857143044471741,
-      "eval_loss": 1.0611016750335693,
-      "eval_runtime": 47.5705,
-      "eval_samples_per_second": 1.472,
-      "eval_steps_per_second": 0.378,
-      "step": 1160
-    },
-    {
-      "epoch": 19.99,
-      "step": 1160,
-      "total_flos": 6.655036715585741e+18,
-      "train_loss": 0.6480823286648454,
-      "train_runtime": 27204.8,
-      "train_samples_per_second": 0.515,
-      "train_steps_per_second": 0.043
+      "epoch": 9.0,
+      "eval_accuracy": 0.6428571343421936,
+      "eval_loss": 1.353231430053711,
+      "eval_runtime": 47.1844,
+      "eval_samples_per_second": 1.484,
+      "eval_steps_per_second": 0.381,
+      "step": 1575
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 2.4e-08,
+      "loss": 0.4008,
+      "step": 1750
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.7285714149475098,
+      "eval_loss": 1.1368242502212524,
+      "eval_runtime": 47.4087,
+      "eval_samples_per_second": 1.477,
+      "eval_steps_per_second": 0.38,
+      "step": 1750
+    },
+    {
+      "epoch": 10.0,
+      "step": 1750,
+      "total_flos": 3.328882796061573e+18,
+      "train_loss": 0.7426296255929129,
+      "train_runtime": 13563.0274,
+      "train_samples_per_second": 0.516,
+      "train_steps_per_second": 0.129
     }
   ],
-  "max_steps": 1160,
-  "num_train_epochs": 20,
-  "total_flos": 6.655036715585741e+18,
+  "max_steps": 1750,
+  "num_train_epochs": 10,
+  "total_flos": 3.328882796061573e+18,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:990634455712da4e6301f173f9ebb709ca13e2946d99da7b5ffd9662eec3717f
+oid sha256:5cda600d7e556b18aa0b24e2e84fec73d5e94563f380c420560e3b34a61273f9
 size 3055