End of training

README.md

@@ -26,16 +26,16 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.842443729903537
+      value: 0.8553054662379421
     - name: Precision
       type: precision
-      value: 0.8395802314162165
+      value: 0.8675973805921082
     - name: Recall
       type: recall
-      value: 0.842443729903537
+      value: 0.8553054662379421
     - name: F1
       type: f1
-      value: 0.8393214641540072
+      value: 0.8581712564304036
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -45,11 +45,11 @@ 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: 0.2213
-- Accuracy: 0.8424
-- Precision: 0.8396
-- Recall: 0.8424
-- F1: 0.8393
+- Loss: 0.2816
+- Accuracy: 0.8553
+- Precision: 0.8676
+- Recall: 0.8553
+- F1: 0.8582
 
 ## Model description
 

all_results.json

@@ -1,8 +1,16 @@
 {
     "epoch": 10.0,
+    "eval_accuracy": 0.8553054662379421,
+    "eval_f1": 0.8581712564304036,
+    "eval_loss": 0.2815697491168976,
+    "eval_precision": 0.8675973805921082,
+    "eval_recall": 0.8553054662379421,
+    "eval_runtime": 16.5636,
+    "eval_samples_per_second": 18.776,
+    "eval_steps_per_second": 0.604,
     "total_flos": 6.954705718242509e+17,
-    "train_loss": 0.23402113264257257,
-    "train_runtime": 1621.2514,
-    "train_samples_per_second": 17.258,
-    "train_steps_per_second": 0.136
+    "train_loss": 0.3014452880079096,
+    "train_runtime": 4522.4246,
+    "train_samples_per_second": 6.187,
+    "train_steps_per_second": 0.049
 }

eval_results.json

@@ -1,11 +1,11 @@
 {
-    "epoch": 9.887640449438202,
-    "eval_accuracy": 0.9714285714285714,
-    "eval_f1": 0.9695078031212484,
-    "eval_loss": 0.08143213391304016,
-    "eval_precision": 0.9696825396825397,
-    "eval_recall": 0.9714285714285714,
-    "eval_runtime": 31.2051,
-    "eval_samples_per_second": 10.094,
-    "eval_steps_per_second": 0.32
+    "epoch": 10.0,
+    "eval_accuracy": 0.8553054662379421,
+    "eval_f1": 0.8581712564304036,
+    "eval_loss": 0.2815697491168976,
+    "eval_precision": 0.8675973805921082,
+    "eval_recall": 0.8553054662379421,
+    "eval_runtime": 16.5636,
+    "eval_samples_per_second": 18.776,
+    "eval_steps_per_second": 0.604
 }

runs/Oct30_11-14-51_d615b0b3af7a/events.out.tfevents.1730291607.d615b0b3af7a.443.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6a0a1830d870a2630da3e6988c0bbe8ebcb86b0f810153fa779b4720f00dcb88
+size 560

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 10.0,
     "total_flos": 6.954705718242509e+17,
-    "train_loss": 0.23402113264257257,
-    "train_runtime": 1621.2514,
-    "train_samples_per_second": 17.258,
-    "train_steps_per_second": 0.136
+    "train_loss": 0.3014452880079096,
+    "train_runtime": 4522.4246,
+    "train_samples_per_second": 6.187,
+    "train_steps_per_second": 0.049
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.8866264991906878,
-  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-220",
+  "best_metric": 0.8581712564304036,
+  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-66",
   "epoch": 10.0,
   "eval_steps": 500,
   "global_step": 220,
@@ -10,286 +10,286 @@
   "log_history": [
     {
       "epoch": 0.45454545454545453,
-      "grad_norm": 3.2964470386505127,
+      "grad_norm": 6.432889938354492,
       "learning_rate": 2.272727272727273e-05,
-      "loss": 0.2322,
+      "loss": 0.6765,
       "step": 10
     },
     {
       "epoch": 0.9090909090909091,
-      "grad_norm": 2.9043655395507812,
+      "grad_norm": 8.764992713928223,
       "learning_rate": 4.545454545454546e-05,
-      "loss": 0.2604,
+      "loss": 0.5793,
       "step": 20
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.8392282958199357,
-      "eval_f1": 0.8332760485475542,
-      "eval_loss": 0.2333919107913971,
-      "eval_precision": 0.8374611693960878,
-      "eval_recall": 0.8392282958199357,
-      "eval_runtime": 15.9334,
-      "eval_samples_per_second": 19.519,
-      "eval_steps_per_second": 0.628,
+      "eval_accuracy": 0.6784565916398714,
+      "eval_f1": 0.5484840645057965,
+      "eval_loss": 0.5874345302581787,
+      "eval_precision": 0.4603033467395912,
+      "eval_recall": 0.6784565916398714,
+      "eval_runtime": 293.1777,
+      "eval_samples_per_second": 1.061,
+      "eval_steps_per_second": 0.034,
       "step": 22
     },
     {
       "epoch": 1.3636363636363638,
-      "grad_norm": 56.99660873413086,
+      "grad_norm": 7.319944381713867,
       "learning_rate": 4.797979797979798e-05,
-      "loss": 0.2223,
+      "loss": 0.4587,
       "step": 30
     },
     {
       "epoch": 1.8181818181818183,
-      "grad_norm": 18.48450469970703,
+      "grad_norm": 11.48066520690918,
       "learning_rate": 4.545454545454546e-05,
-      "loss": 0.2492,
+      "loss": 0.3711,
       "step": 40
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8263665594855305,
-      "eval_f1": 0.8232255290144138,
-      "eval_loss": 0.2620704472064972,
-      "eval_precision": 0.8228686443399579,
-      "eval_recall": 0.8263665594855305,
-      "eval_runtime": 15.7143,
-      "eval_samples_per_second": 19.791,
-      "eval_steps_per_second": 0.636,
+      "eval_accuracy": 0.7781350482315113,
+      "eval_f1": 0.7395292980112433,
+      "eval_loss": 0.4134717881679535,
+      "eval_precision": 0.8169282951542143,
+      "eval_recall": 0.7781350482315113,
+      "eval_runtime": 15.8701,
+      "eval_samples_per_second": 19.597,
+      "eval_steps_per_second": 0.63,
       "step": 44
     },
     {
       "epoch": 2.2727272727272725,
-      "grad_norm": 6.249089241027832,
+      "grad_norm": 36.77213668823242,
       "learning_rate": 4.292929292929293e-05,
-      "loss": 0.2374,
+      "loss": 0.3655,
       "step": 50
     },
     {
       "epoch": 2.7272727272727275,
-      "grad_norm": 34.324771881103516,
+      "grad_norm": 14.163016319274902,
       "learning_rate": 4.0404040404040405e-05,
-      "loss": 0.2373,
+      "loss": 0.2961,
       "step": 60
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.8617363344051447,
-      "eval_f1": 0.8625935578100142,
-      "eval_loss": 0.23337994515895844,
-      "eval_precision": 0.8639282703997218,
-      "eval_recall": 0.8617363344051447,
-      "eval_runtime": 16.0907,
-      "eval_samples_per_second": 19.328,
-      "eval_steps_per_second": 0.621,
+      "eval_accuracy": 0.8553054662379421,
+      "eval_f1": 0.8581712564304036,
+      "eval_loss": 0.2815697491168976,
+      "eval_precision": 0.8675973805921082,
+      "eval_recall": 0.8553054662379421,
+      "eval_runtime": 16.8336,
+      "eval_samples_per_second": 18.475,
+      "eval_steps_per_second": 0.594,
       "step": 66
     },
     {
       "epoch": 3.1818181818181817,
-      "grad_norm": 12.280445098876953,
+      "grad_norm": 10.475957870483398,
       "learning_rate": 3.787878787878788e-05,
-      "loss": 0.2634,
+      "loss": 0.2839,
       "step": 70
     },
     {
       "epoch": 3.6363636363636362,
-      "grad_norm": 4.722353935241699,
+      "grad_norm": 7.20914888381958,
       "learning_rate": 3.535353535353535e-05,
-      "loss": 0.2623,
+      "loss": 0.2576,
       "step": 80
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.8810289389067524,
-      "eval_f1": 0.88319217217326,
-      "eval_loss": 0.2036450058221817,
-      "eval_precision": 0.8908256857340293,
-      "eval_recall": 0.8810289389067524,
-      "eval_runtime": 16.3075,
-      "eval_samples_per_second": 19.071,
-      "eval_steps_per_second": 0.613,
+      "eval_accuracy": 0.7942122186495176,
+      "eval_f1": 0.7857181245354792,
+      "eval_loss": 0.2898705303668976,
+      "eval_precision": 0.7883531432398468,
+      "eval_recall": 0.7942122186495176,
+      "eval_runtime": 16.0846,
+      "eval_samples_per_second": 19.335,
+      "eval_steps_per_second": 0.622,
       "step": 88
     },
     {
       "epoch": 4.090909090909091,
-      "grad_norm": 11.214787483215332,
+      "grad_norm": 8.099700927734375,
       "learning_rate": 3.282828282828283e-05,
-      "loss": 0.2246,
+      "loss": 0.2458,
       "step": 90
     },
     {
       "epoch": 4.545454545454545,
-      "grad_norm": 13.322309494018555,
+      "grad_norm": 16.755117416381836,
       "learning_rate": 3.0303030303030306e-05,
-      "loss": 0.232,
+      "loss": 0.2473,
       "step": 100
     },
     {
       "epoch": 5.0,
-      "grad_norm": 8.625347137451172,
+      "grad_norm": 14.67806339263916,
       "learning_rate": 2.777777777777778e-05,
-      "loss": 0.2378,
+      "loss": 0.261,
       "step": 110
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.819935691318328,
-      "eval_f1": 0.8172812345261152,
-      "eval_loss": 0.2944244146347046,
-      "eval_precision": 0.8165197977946229,
-      "eval_recall": 0.819935691318328,
-      "eval_runtime": 16.454,
-      "eval_samples_per_second": 18.901,
-      "eval_steps_per_second": 0.608,
+      "eval_accuracy": 0.8102893890675241,
+      "eval_f1": 0.8036576399029299,
+      "eval_loss": 0.24693283438682556,
+      "eval_precision": 0.8056915007838585,
+      "eval_recall": 0.8102893890675241,
+      "eval_runtime": 16.5305,
+      "eval_samples_per_second": 18.814,
+      "eval_steps_per_second": 0.605,
       "step": 110
     },
     {
       "epoch": 5.454545454545454,
-      "grad_norm": 5.456847667694092,
+      "grad_norm": 19.368623733520508,
       "learning_rate": 2.5252525252525256e-05,
-      "loss": 0.2711,
+      "loss": 0.2877,
       "step": 120
     },
     {
       "epoch": 5.909090909090909,
-      "grad_norm": 11.77868938446045,
+      "grad_norm": 5.459014892578125,
       "learning_rate": 2.272727272727273e-05,
-      "loss": 0.221,
+      "loss": 0.2559,
       "step": 130
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.8681672025723473,
-      "eval_f1": 0.8700951269031074,
-      "eval_loss": 0.20269730687141418,
-      "eval_precision": 0.8751993658233878,
-      "eval_recall": 0.8681672025723473,
-      "eval_runtime": 15.757,
-      "eval_samples_per_second": 19.737,
-      "eval_steps_per_second": 0.635,
+      "eval_accuracy": 0.8360128617363344,
+      "eval_f1": 0.8179357935850551,
+      "eval_loss": 0.25484344363212585,
+      "eval_precision": 0.8632228058178503,
+      "eval_recall": 0.8360128617363344,
+      "eval_runtime": 16.2848,
+      "eval_samples_per_second": 19.098,
+      "eval_steps_per_second": 0.614,
       "step": 132
     },
     {
       "epoch": 6.363636363636363,
-      "grad_norm": 10.365585327148438,
+      "grad_norm": 3.7228970527648926,
       "learning_rate": 2.0202020202020203e-05,
-      "loss": 0.2205,
+      "loss": 0.2382,
       "step": 140
     },
     {
       "epoch": 6.818181818181818,
-      "grad_norm": 3.4123003482818604,
+      "grad_norm": 7.972034454345703,
       "learning_rate": 1.7676767676767676e-05,
-      "loss": 0.2339,
+      "loss": 0.2249,
       "step": 150
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.8585209003215434,
-      "eval_f1": 0.8577414571643612,
-      "eval_loss": 0.2290601134300232,
-      "eval_precision": 0.8572771006256387,
-      "eval_recall": 0.8585209003215434,
-      "eval_runtime": 15.7364,
-      "eval_samples_per_second": 19.763,
-      "eval_steps_per_second": 0.635,
+      "eval_accuracy": 0.8135048231511254,
+      "eval_f1": 0.7881662757619652,
+      "eval_loss": 0.2835286557674408,
+      "eval_precision": 0.8478531476110869,
+      "eval_recall": 0.8135048231511254,
+      "eval_runtime": 17.1131,
+      "eval_samples_per_second": 18.173,
+      "eval_steps_per_second": 0.584,
       "step": 154
     },
     {
       "epoch": 7.2727272727272725,
-      "grad_norm": 6.866532325744629,
+      "grad_norm": 4.717079162597656,
       "learning_rate": 1.5151515151515153e-05,
-      "loss": 0.2274,
+      "loss": 0.2424,
       "step": 160
     },
     {
       "epoch": 7.7272727272727275,
-      "grad_norm": 28.103683471679688,
+      "grad_norm": 3.11995267868042,
       "learning_rate": 1.2626262626262628e-05,
-      "loss": 0.2215,
+      "loss": 0.2242,
       "step": 170
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.8681672025723473,
-      "eval_f1": 0.868338573836183,
-      "eval_loss": 0.2732490599155426,
-      "eval_precision": 0.8685295263564022,
-      "eval_recall": 0.8681672025723473,
-      "eval_runtime": 17.2464,
-      "eval_samples_per_second": 18.033,
-      "eval_steps_per_second": 0.58,
+      "eval_accuracy": 0.8295819935691319,
+      "eval_f1": 0.8262048489829058,
+      "eval_loss": 0.23345188796520233,
+      "eval_precision": 0.8261307807855373,
+      "eval_recall": 0.8295819935691319,
+      "eval_runtime": 17.7042,
+      "eval_samples_per_second": 17.566,
+      "eval_steps_per_second": 0.565,
       "step": 176
     },
     {
       "epoch": 8.181818181818182,
-      "grad_norm": 7.824560642242432,
+      "grad_norm": 8.103911399841309,
       "learning_rate": 1.0101010101010101e-05,
-      "loss": 0.2202,
+      "loss": 0.229,
       "step": 180
     },
     {
       "epoch": 8.636363636363637,
-      "grad_norm": 1.545398473739624,
+      "grad_norm": 5.379126071929932,
       "learning_rate": 7.5757575757575764e-06,
-      "loss": 0.2162,
+      "loss": 0.2215,
       "step": 190
     },
     {
       "epoch": 9.0,
-      "eval_accuracy": 0.8681672025723473,
-      "eval_f1": 0.8692844166638126,
-      "eval_loss": 0.2260492444038391,
-      "eval_precision": 0.8713400647093587,
-      "eval_recall": 0.8681672025723473,
-      "eval_runtime": 15.7507,
-      "eval_samples_per_second": 19.745,
-      "eval_steps_per_second": 0.635,
+      "eval_accuracy": 0.8520900321543409,
+      "eval_f1": 0.8531774023903234,
+      "eval_loss": 0.22933121025562286,
+      "eval_precision": 0.8549399751110901,
+      "eval_recall": 0.8520900321543409,
+      "eval_runtime": 16.8963,
+      "eval_samples_per_second": 18.406,
+      "eval_steps_per_second": 0.592,
       "step": 198
     },
     {
       "epoch": 9.090909090909092,
-      "grad_norm": 2.2911288738250732,
+      "grad_norm": 18.202444076538086,
       "learning_rate": 5.050505050505051e-06,
-      "loss": 0.2224,
+      "loss": 0.2223,
       "step": 200
     },
     {
       "epoch": 9.545454545454545,
-      "grad_norm": 3.872638463973999,
+      "grad_norm": 2.7657723426818848,
       "learning_rate": 2.5252525252525253e-06,
-      "loss": 0.2126,
+      "loss": 0.2159,
       "step": 210
     },
     {
       "epoch": 10.0,
-      "grad_norm": 2.078859567642212,
+      "grad_norm": 8.540946006774902,
       "learning_rate": 0.0,
-      "loss": 0.2226,
+      "loss": 0.2269,
       "step": 220
     },
     {
       "epoch": 10.0,
-      "eval_accuracy": 0.8842443729903537,
-      "eval_f1": 0.8866264991906878,
-      "eval_loss": 0.20013029873371124,
-      "eval_precision": 0.8965561746996403,
-      "eval_recall": 0.8842443729903537,
-      "eval_runtime": 16.6583,
-      "eval_samples_per_second": 18.669,
-      "eval_steps_per_second": 0.6,
+      "eval_accuracy": 0.842443729903537,
+      "eval_f1": 0.8393214641540072,
+      "eval_loss": 0.2212739735841751,
+      "eval_precision": 0.8395802314162165,
+      "eval_recall": 0.842443729903537,
+      "eval_runtime": 16.6465,
+      "eval_samples_per_second": 18.683,
+      "eval_steps_per_second": 0.601,
       "step": 220
     },
     {
       "epoch": 10.0,
       "step": 220,
       "total_flos": 6.954705718242509e+17,
-      "train_loss": 0.23402113264257257,
-      "train_runtime": 1621.2514,
-      "train_samples_per_second": 17.258,
-      "train_steps_per_second": 0.136
+      "train_loss": 0.3014452880079096,
+      "train_runtime": 4522.4246,
+      "train_samples_per_second": 6.187,
+      "train_steps_per_second": 0.049
     }
   ],
   "logging_steps": 10,