End of training

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 8.11,
-    "eval_accuracy": 0.0,
-    "eval_loss": NaN,
-    "eval_runtime": 5.2931,
-    "eval_samples_per_second": 55.733,
-    "eval_steps_per_second": 1.889,
-    "total_flos": 3.675135085585367e+17,
-    "train_loss": 0.0,
-    "train_runtime": 249.2359,
-    "train_samples_per_second": 47.345,
-    "train_steps_per_second": 0.12
+    "eval_accuracy": 0.8542372881355932,
+    "eval_loss": 0.5490496158599854,
+    "eval_runtime": 29.4646,
+    "eval_samples_per_second": 10.012,
+    "eval_steps_per_second": 0.339,
+    "total_flos": 3.673187375325512e+17,
+    "train_loss": 0.8043342590332031,
+    "train_runtime": 2700.5846,
+    "train_samples_per_second": 4.369,
+    "train_steps_per_second": 0.011
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 8.11,
-    "eval_accuracy": 0.0,
-    "eval_loss": NaN,
-    "eval_runtime": 5.2931,
-    "eval_samples_per_second": 55.733,
-    "eval_steps_per_second": 1.889
+    "eval_accuracy": 0.8542372881355932,
+    "eval_loss": 0.5490496158599854,
+    "eval_runtime": 29.4646,
+    "eval_samples_per_second": 10.012,
+    "eval_steps_per_second": 0.339
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 8.11,
-    "total_flos": 3.675135085585367e+17,
-    "train_loss": 0.0,
-    "train_runtime": 249.2359,
-    "train_samples_per_second": 47.345,
-    "train_steps_per_second": 0.12
+    "total_flos": 3.673187375325512e+17,
+    "train_loss": 0.8043342590332031,
+    "train_runtime": 2700.5846,
+    "train_samples_per_second": 4.369,
+    "train_steps_per_second": 0.011
 }

trainer_state.json

@@ -1,8 +1,7 @@
 {
-  "best_metric": 0.0,
-  "best_model_checkpoint": "resnet-101-finetuned-CivilEng11k/checkpoint-3",
+  "best_metric": 0.8542372881355932,
+  "best_model_checkpoint": "resnet-101-finetuned-CivilEng11k\\checkpoint-30",
   "epoch": 8.108108108108109,
-  "eval_steps": 500,
   "global_step": 30,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
@@ -10,123 +9,116 @@
   "log_history": [
     {
       "epoch": 0.81,
-      "eval_accuracy": 0.0,
-      "eval_loss": NaN,
-      "eval_runtime": 5.4471,
-      "eval_samples_per_second": 54.157,
-      "eval_steps_per_second": 1.836,
+      "eval_accuracy": 0.5728813559322034,
+      "eval_loss": 1.072370171546936,
+      "eval_runtime": 31.0882,
+      "eval_samples_per_second": 9.489,
+      "eval_steps_per_second": 0.322,
       "step": 3
     },
     {
       "epoch": 1.89,
-      "eval_accuracy": 0.0,
-      "eval_loss": NaN,
-      "eval_runtime": 5.2566,
-      "eval_samples_per_second": 56.12,
-      "eval_steps_per_second": 1.902,
+      "eval_accuracy": 0.6542372881355932,
+      "eval_loss": 0.9717434048652649,
+      "eval_runtime": 28.8812,
+      "eval_samples_per_second": 10.214,
+      "eval_steps_per_second": 0.346,
       "step": 7
     },
     {
       "epoch": 2.7,
-      "grad_norm": NaN,
       "learning_rate": 0.00022222222222222218,
-      "loss": 0.0,
+      "loss": 1.0293,
       "step": 10
     },
     {
       "epoch": 2.97,
-      "eval_accuracy": 0.0,
-      "eval_loss": NaN,
-      "eval_runtime": 5.2368,
-      "eval_samples_per_second": 56.332,
-      "eval_steps_per_second": 1.91,
+      "eval_accuracy": 0.6677966101694915,
+      "eval_loss": 0.8594140410423279,
+      "eval_runtime": 29.6429,
+      "eval_samples_per_second": 9.952,
+      "eval_steps_per_second": 0.337,
       "step": 11
     },
     {
       "epoch": 3.78,
-      "eval_accuracy": 0.0,
-      "eval_loss": NaN,
-      "eval_runtime": 5.1779,
-      "eval_samples_per_second": 56.973,
-      "eval_steps_per_second": 1.931,
+      "eval_accuracy": 0.7016949152542373,
+      "eval_loss": 0.7830201983451843,
+      "eval_runtime": 27.8913,
+      "eval_samples_per_second": 10.577,
+      "eval_steps_per_second": 0.359,
       "step": 14
     },
     {
       "epoch": 4.86,
-      "eval_accuracy": 0.0,
-      "eval_loss": NaN,
-      "eval_runtime": 5.1256,
-      "eval_samples_per_second": 57.555,
-      "eval_steps_per_second": 1.951,
+      "eval_accuracy": 0.7593220338983051,
+      "eval_loss": 0.6764189600944519,
+      "eval_runtime": 28.0114,
+      "eval_samples_per_second": 10.531,
+      "eval_steps_per_second": 0.357,
       "step": 18
     },
     {
       "epoch": 5.41,
-      "grad_norm": NaN,
       "learning_rate": 0.00011111111111111109,
-      "loss": 0.0,
+      "loss": 0.78,
       "step": 20
     },
     {
       "epoch": 5.95,
-      "eval_accuracy": 0.0,
-      "eval_loss": NaN,
-      "eval_runtime": 5.1976,
-      "eval_samples_per_second": 56.757,
-      "eval_steps_per_second": 1.924,
+      "eval_accuracy": 0.7830508474576271,
+      "eval_loss": 0.6071953177452087,
+      "eval_runtime": 27.9144,
+      "eval_samples_per_second": 10.568,
+      "eval_steps_per_second": 0.358,
       "step": 22
     },
     {
       "epoch": 6.76,
-      "eval_accuracy": 0.0,
-      "eval_loss": NaN,
-      "eval_runtime": 5.1726,
-      "eval_samples_per_second": 57.032,
-      "eval_steps_per_second": 1.933,
+      "eval_accuracy": 0.8338983050847457,
+      "eval_loss": 0.5745152235031128,
+      "eval_runtime": 28.2316,
+      "eval_samples_per_second": 10.449,
+      "eval_steps_per_second": 0.354,
       "step": 25
     },
     {
       "epoch": 7.84,
-      "eval_accuracy": 0.0,
-      "eval_loss": NaN,
-      "eval_runtime": 5.0794,
-      "eval_samples_per_second": 58.078,
-      "eval_steps_per_second": 1.969,
+      "eval_accuracy": 0.8508474576271187,
+      "eval_loss": 0.5488777756690979,
+      "eval_runtime": 29.7841,
+      "eval_samples_per_second": 9.905,
+      "eval_steps_per_second": 0.336,
       "step": 29
     },
     {
       "epoch": 8.11,
-      "grad_norm": NaN,
       "learning_rate": 0.0,
-      "loss": 0.0,
+      "loss": 0.6037,
       "step": 30
     },
     {
       "epoch": 8.11,
-      "eval_accuracy": 0.0,
-      "eval_loss": NaN,
-      "eval_runtime": 5.2939,
-      "eval_samples_per_second": 55.725,
-      "eval_steps_per_second": 1.889,
+      "eval_accuracy": 0.8542372881355932,
+      "eval_loss": 0.5490496158599854,
+      "eval_runtime": 29.9582,
+      "eval_samples_per_second": 9.847,
+      "eval_steps_per_second": 0.334,
       "step": 30
     },
     {
       "epoch": 8.11,
       "step": 30,
-      "total_flos": 3.675135085585367e+17,
-      "train_loss": 0.0,
-      "train_runtime": 249.2359,
-      "train_samples_per_second": 47.345,
-      "train_steps_per_second": 0.12
+      "total_flos": 3.673187375325512e+17,
+      "train_loss": 0.8043342590332031,
+      "train_runtime": 2700.5846,
+      "train_samples_per_second": 4.369,
+      "train_steps_per_second": 0.011
     }
   ],
-  "logging_steps": 10,
   "max_steps": 30,
-  "num_input_tokens_seen": 0,
   "num_train_epochs": 10,
-  "save_steps": 500,
-  "total_flos": 3.675135085585367e+17,
-  "train_batch_size": 32,
+  "total_flos": 3.673187375325512e+17,
   "trial_name": null,
   "trial_params": null
 }