End of training

README.md

@@ -22,7 +22,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.9019607843137255
+      value: 0.9052287581699346
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -32,8 +32,8 @@ 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: 0.2269
-- Accuracy: 0.9020
+- Loss: 0.2288
+- Accuracy: 0.9052
 
 ## Model description
 

all_results.json

@@ -1,13 +1,13 @@
 {
-    "epoch": 2.79,
+    "epoch": 3.72,
     "eval_accuracy": 0.9052287581699346,
-    "eval_loss": 0.23792558908462524,
-    "eval_runtime": 2.1026,
-    "eval_samples_per_second": 145.535,
-    "eval_steps_per_second": 2.378,
-    "total_flos": 1.9084961608228454e+17,
-    "train_loss": 0.389277982711792,
-    "train_runtime": 162.1827,
-    "train_samples_per_second": 50.887,
-    "train_steps_per_second": 0.185
+    "eval_loss": 0.22878025472164154,
+    "eval_runtime": 2.2699,
+    "eval_samples_per_second": 134.807,
+    "eval_steps_per_second": 2.203,
+    "total_flos": 2.5445786921520538e+17,
+    "train_loss": 0.1845790594816208,
+    "train_runtime": 114.4495,
+    "train_samples_per_second": 96.147,
+    "train_steps_per_second": 0.349
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 2.79,
+    "epoch": 3.72,
     "eval_accuracy": 0.9052287581699346,
-    "eval_loss": 0.23792558908462524,
-    "eval_runtime": 2.1026,
-    "eval_samples_per_second": 145.535,
-    "eval_steps_per_second": 2.378
+    "eval_loss": 0.22878025472164154,
+    "eval_runtime": 2.2699,
+    "eval_samples_per_second": 134.807,
+    "eval_steps_per_second": 2.203
 }

runs/May10_21-58-40_223f6404b016/events.out.tfevents.1715378452.223f6404b016.34.3

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:01dcadaba42b43391879dcd21fe57a3cc23ede3883d40f3bb2c845ec22e20a62
+size 405

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 2.79,
-    "total_flos": 1.9084961608228454e+17,
-    "train_loss": 0.389277982711792,
-    "train_runtime": 162.1827,
-    "train_samples_per_second": 50.887,
-    "train_steps_per_second": 0.185
+    "epoch": 3.72,
+    "total_flos": 2.5445786921520538e+17,
+    "train_loss": 0.1845790594816208,
+    "train_runtime": 114.4495,
+    "train_samples_per_second": 96.147,
+    "train_steps_per_second": 0.349
 }

trainer_state.json

@@ -1,77 +1,93 @@
 {
   "best_metric": 0.9052287581699346,
-  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-30",
-  "epoch": 2.7906976744186047,
+  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-10",
+  "epoch": 3.7209302325581395,
   "eval_steps": 500,
-  "global_step": 30,
+  "global_step": 40,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.93,
-      "grad_norm": 3.6813011169433594,
-      "learning_rate": 3.7037037037037037e-05,
-      "loss": 0.6926,
+      "grad_norm": 4.222448825836182,
+      "learning_rate": 4.166666666666667e-05,
+      "loss": 0.207,
       "step": 10
     },
     {
       "epoch": 0.93,
-      "eval_accuracy": 0.8758169934640523,
-      "eval_loss": 0.3020870089530945,
-      "eval_runtime": 2.9709,
-      "eval_samples_per_second": 102.999,
-      "eval_steps_per_second": 1.683,
+      "eval_accuracy": 0.9052287581699346,
+      "eval_loss": 0.22878025472164154,
+      "eval_runtime": 1.601,
+      "eval_samples_per_second": 191.126,
+      "eval_steps_per_second": 3.123,
       "step": 10
     },
     {
       "epoch": 1.86,
-      "grad_norm": 4.077437877655029,
-      "learning_rate": 1.8518518518518518e-05,
-      "loss": 0.2563,
+      "grad_norm": 3.6448166370391846,
+      "learning_rate": 2.777777777777778e-05,
+      "loss": 0.1873,
       "step": 20
     },
     {
       "epoch": 1.95,
       "eval_accuracy": 0.8954248366013072,
-      "eval_loss": 0.2383614331483841,
-      "eval_runtime": 1.6079,
-      "eval_samples_per_second": 190.312,
-      "eval_steps_per_second": 3.11,
+      "eval_loss": 0.24186921119689941,
+      "eval_runtime": 1.5552,
+      "eval_samples_per_second": 196.755,
+      "eval_steps_per_second": 3.215,
       "step": 21
     },
     {
       "epoch": 2.79,
-      "grad_norm": 3.6792750358581543,
-      "learning_rate": 0.0,
-      "loss": 0.219,
+      "grad_norm": 3.7166202068328857,
+      "learning_rate": 1.388888888888889e-05,
+      "loss": 0.1808,
       "step": 30
     },
     {
-      "epoch": 2.79,
+      "epoch": 2.98,
       "eval_accuracy": 0.9052287581699346,
-      "eval_loss": 0.23792558908462524,
-      "eval_runtime": 1.697,
-      "eval_samples_per_second": 180.314,
-      "eval_steps_per_second": 2.946,
-      "step": 30
+      "eval_loss": 0.23222029209136963,
+      "eval_runtime": 1.6942,
+      "eval_samples_per_second": 180.612,
+      "eval_steps_per_second": 2.951,
+      "step": 32
     },
     {
-      "epoch": 2.79,
-      "step": 30,
-      "total_flos": 1.9084961608228454e+17,
-      "train_loss": 0.389277982711792,
-      "train_runtime": 162.1827,
-      "train_samples_per_second": 50.887,
-      "train_steps_per_second": 0.185
+      "epoch": 3.72,
+      "grad_norm": 5.479237079620361,
+      "learning_rate": 0.0,
+      "loss": 0.1632,
+      "step": 40
+    },
+    {
+      "epoch": 3.72,
+      "eval_accuracy": 0.9019607843137255,
+      "eval_loss": 0.2268960028886795,
+      "eval_runtime": 1.7412,
+      "eval_samples_per_second": 175.742,
+      "eval_steps_per_second": 2.872,
+      "step": 40
+    },
+    {
+      "epoch": 3.72,
+      "step": 40,
+      "total_flos": 2.5445786921520538e+17,
+      "train_loss": 0.1845790594816208,
+      "train_runtime": 114.4495,
+      "train_samples_per_second": 96.147,
+      "train_steps_per_second": 0.349
     }
   ],
   "logging_steps": 10,
-  "max_steps": 30,
+  "max_steps": 40,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 3,
+  "num_train_epochs": 4,
   "save_steps": 500,
-  "total_flos": 1.9084961608228454e+17,
+  "total_flos": 2.5445786921520538e+17,
   "train_batch_size": 64,
   "trial_name": null,
   "trial_params": null