End of training

README.md

@@ -33,7 +33,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [microsoft/swin-tiny-patch4-window7-224](https://huggingface.co/microsoft/swin-tiny-patch4-window7-224) on the imagefolder dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0374
+- Loss: 0.0764
 - Accuracy: 1.0
 
 ## Model description

all_results.json

@@ -1,13 +1,13 @@
 {
-    "epoch": 2.769230769230769,
-    "eval_accuracy": 0.9772727272727273,
-    "eval_loss": 0.1939457803964615,
-    "eval_runtime": 1.7666,
-    "eval_samples_per_second": 24.906,
-    "eval_steps_per_second": 1.132,
-    "total_flos": 2.7490723818356736e+16,
-    "train_loss": 0.3741150697072347,
-    "train_runtime": 1085.5289,
-    "train_samples_per_second": 1.086,
-    "train_steps_per_second": 0.008
+    "epoch": 4.615384615384615,
+    "eval_accuracy": 1.0,
+    "eval_loss": 0.07640355080366135,
+    "eval_runtime": 5.0834,
+    "eval_samples_per_second": 8.656,
+    "eval_steps_per_second": 0.393,
+    "total_flos": 4.543674786614477e+16,
+    "train_loss": 0.2573451101779938,
+    "train_runtime": 402.4711,
+    "train_samples_per_second": 4.882,
+    "train_steps_per_second": 0.037
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 2.769230769230769,
-    "eval_accuracy": 0.9772727272727273,
-    "eval_loss": 0.1939457803964615,
-    "eval_runtime": 1.7666,
-    "eval_samples_per_second": 24.906,
-    "eval_steps_per_second": 1.132
+    "epoch": 4.615384615384615,
+    "eval_accuracy": 1.0,
+    "eval_loss": 0.07640355080366135,
+    "eval_runtime": 5.0834,
+    "eval_samples_per_second": 8.656,
+    "eval_steps_per_second": 0.393
 }

runs/Oct24_07-24-26_7919de6d5c0a/events.out.tfevents.1729755160.7919de6d5c0a.285.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7e64e77a9f431ed4a40d1109ce3746748c6329f08fdb6be8726943c89631f875
+size 405

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 2.769230769230769,
-    "total_flos": 2.7490723818356736e+16,
-    "train_loss": 0.3741150697072347,
-    "train_runtime": 1085.5289,
-    "train_samples_per_second": 1.086,
-    "train_steps_per_second": 0.008
+    "epoch": 4.615384615384615,
+    "total_flos": 4.543674786614477e+16,
+    "train_loss": 0.2573451101779938,
+    "train_runtime": 402.4711,
+    "train_samples_per_second": 4.882,
+    "train_steps_per_second": 0.037
 }

trainer_state.json

@@ -1,54 +1,79 @@
 {
-  "best_metric": 0.9772727272727273,
+  "best_metric": 1.0,
   "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-9",
-  "epoch": 2.769230769230769,
+  "epoch": 4.615384615384615,
   "eval_steps": 500,
-  "global_step": 9,
+  "global_step": 15,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.9230769230769231,
-      "eval_accuracy": 0.6136363636363636,
-      "eval_loss": 0.4375806748867035,
-      "eval_runtime": 37.6364,
-      "eval_samples_per_second": 1.169,
-      "eval_steps_per_second": 0.053,
+      "eval_accuracy": 0.6590909090909091,
+      "eval_loss": 0.46299847960472107,
+      "eval_runtime": 21.279,
+      "eval_samples_per_second": 2.068,
+      "eval_steps_per_second": 0.094,
       "step": 3
     },
     {
       "epoch": 1.8461538461538463,
       "eval_accuracy": 0.9090909090909091,
-      "eval_loss": 0.2783834636211395,
-      "eval_runtime": 1.7762,
-      "eval_samples_per_second": 24.772,
-      "eval_steps_per_second": 1.126,
+      "eval_loss": 0.24737197160720825,
+      "eval_runtime": 1.7671,
+      "eval_samples_per_second": 24.899,
+      "eval_steps_per_second": 1.132,
       "step": 6
     },
     {
       "epoch": 2.769230769230769,
-      "eval_accuracy": 0.9772727272727273,
-      "eval_loss": 0.1939457803964615,
-      "eval_runtime": 2.5065,
-      "eval_samples_per_second": 17.554,
-      "eval_steps_per_second": 0.798,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.07640355080366135,
+      "eval_runtime": 1.9688,
+      "eval_samples_per_second": 22.349,
+      "eval_steps_per_second": 1.016,
       "step": 9
     },
     {
-      "epoch": 2.769230769230769,
-      "step": 9,
-      "total_flos": 2.7490723818356736e+16,
-      "train_loss": 0.3741150697072347,
-      "train_runtime": 1085.5289,
-      "train_samples_per_second": 1.086,
-      "train_steps_per_second": 0.008
+      "epoch": 3.076923076923077,
+      "grad_norm": 7.906944274902344,
+      "learning_rate": 1.923076923076923e-05,
+      "loss": 0.3656,
+      "step": 10
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.03935603052377701,
+      "eval_runtime": 2.5937,
+      "eval_samples_per_second": 16.965,
+      "eval_steps_per_second": 0.771,
+      "step": 13
+    },
+    {
+      "epoch": 4.615384615384615,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.03737922012805939,
+      "eval_runtime": 3.8031,
+      "eval_samples_per_second": 11.57,
+      "eval_steps_per_second": 0.526,
+      "step": 15
+    },
+    {
+      "epoch": 4.615384615384615,
+      "step": 15,
+      "total_flos": 4.543674786614477e+16,
+      "train_loss": 0.2573451101779938,
+      "train_runtime": 402.4711,
+      "train_samples_per_second": 4.882,
+      "train_steps_per_second": 0.037
     }
   ],
   "logging_steps": 10,
-  "max_steps": 9,
+  "max_steps": 15,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 3,
+  "num_train_epochs": 5,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -62,7 +87,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 2.7490723818356736e+16,
+  "total_flos": 4.543674786614477e+16,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": null