End of training

all_results.json

@@ -1,13 +1,13 @@
 {
-    "epoch": 2.0,
-    "eval_accuracy": 0.8095238095238095,
-    "eval_loss": 0.5368531346321106,
-    "eval_runtime": 0.2448,
-    "eval_samples_per_second": 85.768,
-    "eval_steps_per_second": 4.084,
-    "total_flos": 8997868012879872.0,
-    "train_loss": 0.6559792359670004,
-    "train_runtime": 10.9739,
-    "train_samples_per_second": 49.481,
-    "train_steps_per_second": 0.273
+    "epoch": 2.88,
+    "eval_accuracy": 0.9602272727272727,
+    "eval_loss": 0.09607520699501038,
+    "eval_runtime": 1.4687,
+    "eval_samples_per_second": 119.833,
+    "eval_steps_per_second": 4.085,
+    "total_flos": 1.1334330977550336e+17,
+    "train_loss": 0.3907758593559265,
+    "train_runtime": 85.0457,
+    "train_samples_per_second": 55.594,
+    "train_steps_per_second": 0.423
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 2.0,
-    "eval_accuracy": 0.8095238095238095,
-    "eval_loss": 0.5368531346321106,
-    "eval_runtime": 0.2448,
-    "eval_samples_per_second": 85.768,
-    "eval_steps_per_second": 4.084
+    "epoch": 2.88,
+    "eval_accuracy": 0.9602272727272727,
+    "eval_loss": 0.09607520699501038,
+    "eval_runtime": 1.4687,
+    "eval_samples_per_second": 119.833,
+    "eval_steps_per_second": 4.085
 }

runs/May04_03-41-04_d7a051415e6a/events.out.tfevents.1714794201.d7a051415e6a.664.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:98780022d7b43897038743639c963c687c260461f48098e7a95f5547b6f7a7be
+size 405

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 2.0,
-    "total_flos": 8997868012879872.0,
-    "train_loss": 0.6559792359670004,
-    "train_runtime": 10.9739,
-    "train_samples_per_second": 49.481,
-    "train_steps_per_second": 0.273
+    "epoch": 2.88,
+    "total_flos": 1.1334330977550336e+17,
+    "train_loss": 0.3907758593559265,
+    "train_runtime": 85.0457,
+    "train_samples_per_second": 55.594,
+    "train_steps_per_second": 0.423
 }

trainer_state.json

@@ -1,47 +1,77 @@
 {
-  "best_metric": 0.8095238095238095,
-  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-3",
-  "epoch": 2.0,
+  "best_metric": 0.9602272727272727,
+  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-36",
+  "epoch": 2.88,
   "eval_steps": 500,
-  "global_step": 3,
+  "global_step": 36,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.6666666666666666,
-      "eval_accuracy": 0.5238095238095238,
-      "eval_loss": 0.6702501177787781,
-      "eval_runtime": 0.1764,
-      "eval_samples_per_second": 119.015,
-      "eval_steps_per_second": 5.667,
-      "step": 1
+      "epoch": 0.8,
+      "grad_norm": 5.466047286987305,
+      "learning_rate": 4.0625000000000005e-05,
+      "loss": 0.6516,
+      "step": 10
     },
     {
-      "epoch": 2.0,
-      "eval_accuracy": 0.8095238095238095,
-      "eval_loss": 0.5368531346321106,
-      "eval_runtime": 0.2329,
-      "eval_samples_per_second": 90.166,
-      "eval_steps_per_second": 4.294,
-      "step": 3
+      "epoch": 0.96,
+      "eval_accuracy": 0.9318181818181818,
+      "eval_loss": 0.36171892285346985,
+      "eval_runtime": 1.4119,
+      "eval_samples_per_second": 124.651,
+      "eval_steps_per_second": 4.249,
+      "step": 12
+    },
+    {
+      "epoch": 1.6,
+      "grad_norm": 8.815180778503418,
+      "learning_rate": 2.5e-05,
+      "loss": 0.3909,
+      "step": 20
     },
     {
       "epoch": 2.0,
-      "step": 3,
-      "total_flos": 8997868012879872.0,
-      "train_loss": 0.6559792359670004,
-      "train_runtime": 10.9739,
-      "train_samples_per_second": 49.481,
-      "train_steps_per_second": 0.273
+      "eval_accuracy": 0.9488636363636364,
+      "eval_loss": 0.15098439157009125,
+      "eval_runtime": 1.4258,
+      "eval_samples_per_second": 123.442,
+      "eval_steps_per_second": 4.208,
+      "step": 25
+    },
+    {
+      "epoch": 2.4,
+      "grad_norm": 7.026189804077148,
+      "learning_rate": 9.375000000000001e-06,
+      "loss": 0.2274,
+      "step": 30
+    },
+    {
+      "epoch": 2.88,
+      "eval_accuracy": 0.9602272727272727,
+      "eval_loss": 0.09607520699501038,
+      "eval_runtime": 1.5278,
+      "eval_samples_per_second": 115.199,
+      "eval_steps_per_second": 3.927,
+      "step": 36
+    },
+    {
+      "epoch": 2.88,
+      "step": 36,
+      "total_flos": 1.1334330977550336e+17,
+      "train_loss": 0.3907758593559265,
+      "train_runtime": 85.0457,
+      "train_samples_per_second": 55.594,
+      "train_steps_per_second": 0.423
     }
   ],
   "logging_steps": 10,
-  "max_steps": 3,
+  "max_steps": 36,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 3,
   "save_steps": 500,
-  "total_flos": 8997868012879872.0,
+  "total_flos": 1.1334330977550336e+17,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": null