End of training

all_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 10.0,
     "train_loss": 0.6484309974533823,
-    "train_runtime": 10912.4775,
+    "train_runtime": 11384.9561,
     "train_samples": 100102,
-    "train_samples_per_second": 91.732,
-    "train_steps_per_second": 0.358
+    "train_samples_per_second": 87.925,
+    "train_steps_per_second": 0.343
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 10.0,
     "train_loss": 0.6484309974533823,
-    "train_runtime": 10912.4775,
+    "train_runtime": 11384.9561,
     "train_samples": 100102,
-    "train_samples_per_second": 91.732,
-    "train_steps_per_second": 0.358
+    "train_samples_per_second": 87.925,
+    "train_steps_per_second": 0.343
 }

train_results.txt

@@ -1,6 +1,6 @@
 epoch = 10.0
 train_loss = 0.6484309974533823
-train_runtime = 10912.4775
+train_runtime = 11384.9561
 train_samples = 100102
-train_samples_per_second = 91.732
-train_steps_per_second = 0.358
+train_samples_per_second = 87.925
+train_steps_per_second = 0.343

trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": null,
-  "best_model_checkpoint": null,
+  "best_metric": 0.7961117029190063,
+  "best_model_checkpoint": "./models/emot_and_voice/checkpoint-1173",
   "epoch": 9.999680409076383,
   "global_step": 3910,
   "is_hyper_param_search": false,
@@ -241,6 +241,17 @@
       "loss": 1.0543,
       "step": 390
     },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.6530612244897959,
+      "eval_auprc": 49.690567272400955,
+      "eval_loss": 0.9922736287117004,
+      "eval_micro f1 score": 65.3061224489796,
+      "eval_runtime": 41.8031,
+      "eval_samples_per_second": 266.081,
+      "eval_steps_per_second": 8.325,
+      "step": 391
+    },
     {
       "epoch": 1.02,
       "learning_rate": 1.9948849104859337e-05,
@@ -475,6 +486,17 @@
       "loss": 0.8573,
       "step": 780
     },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.699901105816776,
+      "eval_auprc": 64.40713435079698,
+      "eval_loss": 0.822926938533783,
+      "eval_micro f1 score": 69.9901105816776,
+      "eval_runtime": 41.7698,
+      "eval_samples_per_second": 266.293,
+      "eval_steps_per_second": 8.331,
+      "step": 782
+    },
     {
       "epoch": 2.02,
       "learning_rate": 1.773231031543052e-05,
@@ -709,6 +731,17 @@
       "loss": 0.7217,
       "step": 1170
     },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.7105996583655488,
+      "eval_auprc": 67.46402201308636,
+      "eval_loss": 0.7961117029190063,
+      "eval_micro f1 score": 71.05996583655488,
+      "eval_runtime": 41.7385,
+      "eval_samples_per_second": 266.493,
+      "eval_steps_per_second": 8.338,
+      "step": 1173
+    },
     {
       "epoch": 3.02,
       "learning_rate": 1.5515771526001706e-05,
@@ -943,6 +976,17 @@
       "loss": 0.6305,
       "step": 1560
     },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.7112289849860649,
+      "eval_auprc": 68.51906233767703,
+      "eval_loss": 0.8162872791290283,
+      "eval_micro f1 score": 71.12289849860647,
+      "eval_runtime": 41.7027,
+      "eval_samples_per_second": 266.721,
+      "eval_steps_per_second": 8.345,
+      "step": 1564
+    },
     {
       "epoch": 4.02,
       "learning_rate": 1.329923273657289e-05,
@@ -1177,6 +1221,17 @@
       "loss": 0.5294,
       "step": 1950
     },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.7101501393508945,
+      "eval_auprc": 68.73344594150042,
+      "eval_loss": 0.8205118775367737,
+      "eval_micro f1 score": 71.01501393508946,
+      "eval_runtime": 41.8651,
+      "eval_samples_per_second": 265.687,
+      "eval_steps_per_second": 8.312,
+      "step": 1955
+    },
     {
       "epoch": 5.01,
       "learning_rate": 1.1082693947144077e-05,
@@ -1411,6 +1466,17 @@
       "loss": 0.4689,
       "step": 2340
     },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.7116785040007192,
+      "eval_auprc": 68.77509414907618,
+      "eval_loss": 0.8716110587120056,
+      "eval_micro f1 score": 71.16785040007191,
+      "eval_runtime": 41.7514,
+      "eval_samples_per_second": 266.41,
+      "eval_steps_per_second": 8.335,
+      "step": 2346
+    },
     {
       "epoch": 6.01,
       "learning_rate": 8.86615515771526e-06,
@@ -1645,6 +1711,17 @@
       "loss": 0.433,
       "step": 2730
     },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.7098804279421019,
+      "eval_auprc": 68.36526805926971,
+      "eval_loss": 0.9086081385612488,
+      "eval_micro f1 score": 70.9880427942102,
+      "eval_runtime": 41.7909,
+      "eval_samples_per_second": 266.158,
+      "eval_steps_per_second": 8.327,
+      "step": 2737
+    },
     {
       "epoch": 7.01,
       "learning_rate": 6.649616368286445e-06,
@@ -1879,6 +1956,17 @@
       "loss": 0.419,
       "step": 3120
     },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.7067337948395217,
+      "eval_auprc": 68.46062183107752,
+      "eval_loss": 0.9290213584899902,
+      "eval_micro f1 score": 70.67337948395217,
+      "eval_runtime": 41.7253,
+      "eval_samples_per_second": 266.577,
+      "eval_steps_per_second": 8.34,
+      "step": 3128
+    },
     {
       "epoch": 8.01,
       "learning_rate": 4.43307757885763e-06,
@@ -2113,6 +2201,17 @@
       "loss": 0.3766,
       "step": 3510
     },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.7064640834307291,
+      "eval_auprc": 68.51324967918615,
+      "eval_loss": 0.9618651866912842,
+      "eval_micro f1 score": 70.6464083430729,
+      "eval_runtime": 41.7225,
+      "eval_samples_per_second": 266.595,
+      "eval_steps_per_second": 8.341,
+      "step": 3519
+    },
     {
       "epoch": 9.0,
       "learning_rate": 2.216538789428815e-06,
@@ -2353,14 +2452,25 @@
       "loss": 0.3395,
       "step": 3910
     },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.707183313854176,
+      "eval_auprc": 68.45618753896134,
+      "eval_loss": 0.971839189529419,
+      "eval_micro f1 score": 70.71833138541761,
+      "eval_runtime": 41.6752,
+      "eval_samples_per_second": 266.897,
+      "eval_steps_per_second": 8.35,
+      "step": 3910
+    },
     {
       "epoch": 10.0,
       "step": 3910,
       "total_flos": 1.0186138340339622e+17,
       "train_loss": 0.6484309974533823,
-      "train_runtime": 10912.4775,
-      "train_samples_per_second": 91.732,
-      "train_steps_per_second": 0.358
+      "train_runtime": 11384.9561,
+      "train_samples_per_second": 87.925,
+      "train_steps_per_second": 0.343
     }
   ],
   "max_steps": 3910,