End of training

all_results.json

@@ -1,13 +1,13 @@
 {
-    "epoch": 6.67,
-    "eval_accuracy": 0.6363636363636364,
-    "eval_loss": 0.6085681319236755,
-    "eval_runtime": 0.2829,
-    "eval_samples_per_second": 77.776,
-    "eval_steps_per_second": 3.535,
-    "total_flos": 3.136825810015027e+16,
-    "train_loss": 0.5420236110687255,
-    "train_runtime": 51.4795,
-    "train_samples_per_second": 36.714,
-    "train_steps_per_second": 0.194
+    "epoch": 3.0,
+    "eval_accuracy": 0.7727272727272727,
+    "eval_loss": 0.5769956111907959,
+    "eval_runtime": 0.3139,
+    "eval_samples_per_second": 70.075,
+    "eval_steps_per_second": 6.37,
+    "total_flos": 1.4093345754980352e+16,
+    "train_loss": 0.5933622784084744,
+    "train_runtime": 25.6802,
+    "train_samples_per_second": 22.079,
+    "train_steps_per_second": 0.35
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 6.67,
-    "eval_accuracy": 0.6363636363636364,
-    "eval_loss": 0.6085681319236755,
-    "eval_runtime": 0.2829,
-    "eval_samples_per_second": 77.776,
-    "eval_steps_per_second": 3.535
+    "epoch": 3.0,
+    "eval_accuracy": 0.7727272727272727,
+    "eval_loss": 0.5769956111907959,
+    "eval_runtime": 0.3139,
+    "eval_samples_per_second": 70.075,
+    "eval_steps_per_second": 6.37
 }

runs/Jan11_06-44-02_884473a9f125/events.out.tfevents.1704957927.884473a9f125.5987.11

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bf4cc0459eb3a2dfb9a1799f9e5455b9bdec8b70fbc849304b4407f9adc827c3
+size 405

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 6.67,
-    "total_flos": 3.136825810015027e+16,
-    "train_loss": 0.5420236110687255,
-    "train_runtime": 51.4795,
-    "train_samples_per_second": 36.714,
-    "train_steps_per_second": 0.194
+    "epoch": 3.0,
+    "total_flos": 1.4093345754980352e+16,
+    "train_loss": 0.5933622784084744,
+    "train_runtime": 25.6802,
+    "train_samples_per_second": 22.079,
+    "train_steps_per_second": 0.35
 }

trainer_state.json

@@ -1,97 +1,55 @@
 {
-  "best_metric": 0.6363636363636364,
-  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat-kornia/checkpoint-1",
-  "epoch": 6.666666666666667,
+  "best_metric": 0.7727272727272727,
+  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat-kornia/checkpoint-9",
+  "epoch": 3.0,
   "eval_steps": 500,
-  "global_step": 10,
+  "global_step": 9,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.67,
-      "eval_accuracy": 0.6363636363636364,
-      "eval_loss": 0.6085681319236755,
-      "eval_runtime": 0.3029,
-      "eval_samples_per_second": 72.64,
-      "eval_steps_per_second": 3.302,
-      "step": 1
-    },
-    {
-      "epoch": 2.0,
-      "eval_accuracy": 0.6363636363636364,
-      "eval_loss": 0.6711809039115906,
-      "eval_runtime": 0.2576,
-      "eval_samples_per_second": 85.402,
-      "eval_steps_per_second": 3.882,
+      "epoch": 1.0,
+      "eval_accuracy": 0.2727272727272727,
+      "eval_loss": 2.8753721714019775,
+      "eval_runtime": 0.2391,
+      "eval_samples_per_second": 91.995,
+      "eval_steps_per_second": 8.363,
       "step": 3
     },
     {
-      "epoch": 2.67,
-      "eval_accuracy": 0.5454545454545454,
-      "eval_loss": 0.722430944442749,
-      "eval_runtime": 0.2475,
-      "eval_samples_per_second": 88.894,
-      "eval_steps_per_second": 4.041,
-      "step": 4
-    },
-    {
-      "epoch": 4.0,
-      "eval_accuracy": 0.5909090909090909,
-      "eval_loss": 0.6421019434928894,
-      "eval_runtime": 0.2467,
-      "eval_samples_per_second": 89.16,
-      "eval_steps_per_second": 4.053,
+      "epoch": 2.0,
+      "eval_accuracy": 0.7272727272727273,
+      "eval_loss": 0.5872368812561035,
+      "eval_runtime": 0.2697,
+      "eval_samples_per_second": 81.585,
+      "eval_steps_per_second": 7.417,
       "step": 6
     },
     {
-      "epoch": 4.67,
-      "eval_accuracy": 0.5454545454545454,
-      "eval_loss": 0.6587771773338318,
-      "eval_runtime": 0.2493,
-      "eval_samples_per_second": 88.241,
-      "eval_steps_per_second": 4.011,
-      "step": 7
-    },
-    {
-      "epoch": 6.0,
-      "eval_accuracy": 0.5454545454545454,
-      "eval_loss": 0.7261324524879456,
-      "eval_runtime": 0.2465,
-      "eval_samples_per_second": 89.265,
-      "eval_steps_per_second": 4.058,
+      "epoch": 3.0,
+      "eval_accuracy": 0.7727272727272727,
+      "eval_loss": 0.5769956111907959,
+      "eval_runtime": 0.2361,
+      "eval_samples_per_second": 93.199,
+      "eval_steps_per_second": 8.473,
       "step": 9
     },
     {
-      "epoch": 6.67,
-      "learning_rate": 0.0,
-      "loss": 0.542,
-      "step": 10
-    },
-    {
-      "epoch": 6.67,
-      "eval_accuracy": 0.5454545454545454,
-      "eval_loss": 0.7230197787284851,
-      "eval_runtime": 0.2401,
-      "eval_samples_per_second": 91.642,
-      "eval_steps_per_second": 4.166,
-      "step": 10
-    },
-    {
-      "epoch": 6.67,
-      "step": 10,
-      "total_flos": 3.136825810015027e+16,
-      "train_loss": 0.5420236110687255,
-      "train_runtime": 51.4795,
-      "train_samples_per_second": 36.714,
-      "train_steps_per_second": 0.194
+      "epoch": 3.0,
+      "step": 9,
+      "total_flos": 1.4093345754980352e+16,
+      "train_loss": 0.5933622784084744,
+      "train_runtime": 25.6802,
+      "train_samples_per_second": 22.079,
+      "train_steps_per_second": 0.35
     }
   ],
   "logging_steps": 10,
-  "max_steps": 10,
-  "num_train_epochs": 10,
+  "max_steps": 9,
+  "num_train_epochs": 3,
   "save_steps": 500,
-  "total_flos": 3.136825810015027e+16,
+  "total_flos": 1.4093345754980352e+16,
   "trial_name": null,
   "trial_params": null
 }