End of training

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 3.0,
-    "eval_accuracy": 0.9781481481481481,
-    "eval_loss": 0.062276002019643784,
-    "eval_runtime": 16.6723,
-    "eval_samples_per_second": 161.946,
-    "eval_steps_per_second": 5.098,
+    "eval_accuracy": 0.9740740740740741,
+    "eval_loss": 0.07933878898620605,
+    "eval_runtime": 16.0127,
+    "eval_samples_per_second": 168.616,
+    "eval_steps_per_second": 5.308,
     "total_flos": 1.8124066505760768e+18,
-    "train_loss": 0.3407971503441794,
-    "train_runtime": 1119.3728,
-    "train_samples_per_second": 65.126,
-    "train_steps_per_second": 0.509
+    "train_loss": 0.32989514710610374,
+    "train_runtime": 1304.6847,
+    "train_samples_per_second": 55.876,
+    "train_steps_per_second": 0.437
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 3.0,
-    "eval_accuracy": 0.9781481481481481,
-    "eval_loss": 0.062276002019643784,
-    "eval_runtime": 16.6723,
-    "eval_samples_per_second": 161.946,
-    "eval_steps_per_second": 5.098
+    "eval_accuracy": 0.9740740740740741,
+    "eval_loss": 0.07933878898620605,
+    "eval_runtime": 16.0127,
+    "eval_samples_per_second": 168.616,
+    "eval_steps_per_second": 5.308
 }

runs/Oct20_09-10-24_f2f37b04cb37/events.out.tfevents.1729417035.f2f37b04cb37.406.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e7b77e154db6f3f7c6f93c6bb4c951b706c65f414f08a66fdfb68e1dd233ac4d
+size 411

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 3.0,
     "total_flos": 1.8124066505760768e+18,
-    "train_loss": 0.3407971503441794,
-    "train_runtime": 1119.3728,
-    "train_samples_per_second": 65.126,
-    "train_steps_per_second": 0.509
+    "train_loss": 0.32989514710610374,
+    "train_runtime": 1304.6847,
+    "train_samples_per_second": 55.876,
+    "train_steps_per_second": 0.437
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.9781481481481481,
+  "best_metric": 0.9740740740740741,
   "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-570",
   "epoch": 3.0,
   "eval_steps": 500,
@@ -10,438 +10,438 @@
   "log_history": [
     {
       "epoch": 0.05263157894736842,
-      "grad_norm": 6.441094398498535,
+      "grad_norm": 5.075918197631836,
       "learning_rate": 8.771929824561403e-06,
-      "loss": 2.2771,
+      "loss": 2.266,
       "step": 10
     },
     {
       "epoch": 0.10526315789473684,
-      "grad_norm": 7.560100078582764,
+      "grad_norm": 5.732299327850342,
       "learning_rate": 1.7543859649122806e-05,
-      "loss": 2.0756,
+      "loss": 2.0713,
       "step": 20
     },
     {
       "epoch": 0.15789473684210525,
-      "grad_norm": 8.593420028686523,
+      "grad_norm": 9.110240936279297,
       "learning_rate": 2.6315789473684212e-05,
-      "loss": 1.6159,
+      "loss": 1.6216,
       "step": 30
     },
     {
       "epoch": 0.21052631578947367,
-      "grad_norm": 15.930532455444336,
+      "grad_norm": 10.104762077331543,
       "learning_rate": 3.508771929824561e-05,
-      "loss": 1.0891,
+      "loss": 0.9728,
       "step": 40
     },
     {
       "epoch": 0.2631578947368421,
-      "grad_norm": 9.635261535644531,
+      "grad_norm": 10.321767807006836,
       "learning_rate": 4.3859649122807014e-05,
-      "loss": 0.7175,
+      "loss": 0.6652,
       "step": 50
     },
     {
       "epoch": 0.3157894736842105,
-      "grad_norm": 29.617084503173828,
+      "grad_norm": 10.895342826843262,
       "learning_rate": 4.970760233918128e-05,
-      "loss": 0.5415,
+      "loss": 0.4921,
       "step": 60
     },
     {
       "epoch": 0.3684210526315789,
-      "grad_norm": 12.889870643615723,
+      "grad_norm": 15.903302192687988,
       "learning_rate": 4.8732943469785574e-05,
-      "loss": 0.3873,
+      "loss": 0.4292,
       "step": 70
     },
     {
       "epoch": 0.42105263157894735,
-      "grad_norm": 28.610116958618164,
+      "grad_norm": 35.4083251953125,
       "learning_rate": 4.7758284600389865e-05,
-      "loss": 0.3935,
+      "loss": 0.3899,
       "step": 80
     },
     {
       "epoch": 0.47368421052631576,
-      "grad_norm": 12.129476547241211,
+      "grad_norm": 17.33992576599121,
       "learning_rate": 4.678362573099415e-05,
-      "loss": 0.407,
+      "loss": 0.3323,
       "step": 90
     },
     {
       "epoch": 0.5263157894736842,
-      "grad_norm": 8.134405136108398,
+      "grad_norm": 16.412843704223633,
       "learning_rate": 4.580896686159844e-05,
-      "loss": 0.403,
+      "loss": 0.3374,
       "step": 100
     },
     {
       "epoch": 0.5789473684210527,
-      "grad_norm": 16.965028762817383,
+      "grad_norm": 12.062039375305176,
       "learning_rate": 4.483430799220273e-05,
-      "loss": 0.3666,
+      "loss": 0.3464,
       "step": 110
     },
     {
       "epoch": 0.631578947368421,
-      "grad_norm": 20.727378845214844,
+      "grad_norm": 13.78882122039795,
       "learning_rate": 4.3859649122807014e-05,
-      "loss": 0.3315,
+      "loss": 0.297,
       "step": 120
     },
     {
       "epoch": 0.6842105263157895,
-      "grad_norm": 10.648832321166992,
+      "grad_norm": 19.42721939086914,
       "learning_rate": 4.2884990253411305e-05,
-      "loss": 0.3185,
+      "loss": 0.2825,
       "step": 130
     },
     {
       "epoch": 0.7368421052631579,
-      "grad_norm": 15.240333557128906,
+      "grad_norm": 7.111037731170654,
       "learning_rate": 4.1910331384015596e-05,
-      "loss": 0.3092,
+      "loss": 0.3188,
       "step": 140
     },
     {
       "epoch": 0.7894736842105263,
-      "grad_norm": 12.595602035522461,
+      "grad_norm": 20.616655349731445,
       "learning_rate": 4.093567251461988e-05,
-      "loss": 0.2781,
+      "loss": 0.2742,
       "step": 150
     },
     {
       "epoch": 0.8421052631578947,
-      "grad_norm": 23.477523803710938,
+      "grad_norm": 20.487638473510742,
       "learning_rate": 3.996101364522417e-05,
-      "loss": 0.3165,
+      "loss": 0.2525,
       "step": 160
     },
     {
       "epoch": 0.8947368421052632,
-      "grad_norm": 12.806583404541016,
+      "grad_norm": 16.18035316467285,
       "learning_rate": 3.898635477582846e-05,
-      "loss": 0.2604,
+      "loss": 0.256,
       "step": 170
     },
     {
       "epoch": 0.9473684210526315,
-      "grad_norm": 12.278985023498535,
+      "grad_norm": 8.112160682678223,
       "learning_rate": 3.8011695906432746e-05,
-      "loss": 0.2268,
+      "loss": 0.2473,
       "step": 180
     },
     {
       "epoch": 1.0,
-      "grad_norm": 16.98476791381836,
+      "grad_norm": 12.733054161071777,
       "learning_rate": 3.7037037037037037e-05,
-      "loss": 0.2614,
+      "loss": 0.2912,
       "step": 190
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.9644444444444444,
-      "eval_loss": 0.11767604947090149,
-      "eval_runtime": 16.3432,
-      "eval_samples_per_second": 165.206,
-      "eval_steps_per_second": 5.201,
+      "eval_accuracy": 0.96,
+      "eval_loss": 0.12840119004249573,
+      "eval_runtime": 16.2205,
+      "eval_samples_per_second": 166.456,
+      "eval_steps_per_second": 5.24,
       "step": 190
     },
     {
       "epoch": 1.0526315789473684,
-      "grad_norm": 22.68017578125,
+      "grad_norm": 8.781696319580078,
       "learning_rate": 3.606237816764133e-05,
-      "loss": 0.2571,
+      "loss": 0.2393,
       "step": 200
     },
     {
       "epoch": 1.1052631578947367,
-      "grad_norm": 8.888700485229492,
+      "grad_norm": 6.2473578453063965,
       "learning_rate": 3.508771929824561e-05,
-      "loss": 0.2413,
+      "loss": 0.211,
       "step": 210
     },
     {
       "epoch": 1.1578947368421053,
-      "grad_norm": 15.368131637573242,
+      "grad_norm": 17.877609252929688,
       "learning_rate": 3.41130604288499e-05,
-      "loss": 0.2676,
+      "loss": 0.2774,
       "step": 220
     },
     {
       "epoch": 1.2105263157894737,
-      "grad_norm": 12.329680442810059,
+      "grad_norm": 12.714591979980469,
       "learning_rate": 3.313840155945419e-05,
-      "loss": 0.223,
+      "loss": 0.2359,
       "step": 230
     },
     {
       "epoch": 1.263157894736842,
-      "grad_norm": 18.98630714416504,
+      "grad_norm": 13.344749450683594,
       "learning_rate": 3.216374269005848e-05,
-      "loss": 0.2213,
+      "loss": 0.1995,
       "step": 240
     },
     {
       "epoch": 1.3157894736842106,
-      "grad_norm": 8.994136810302734,
+      "grad_norm": 11.351320266723633,
       "learning_rate": 3.118908382066277e-05,
-      "loss": 0.2417,
+      "loss": 0.2292,
       "step": 250
     },
     {
       "epoch": 1.368421052631579,
-      "grad_norm": 17.037776947021484,
+      "grad_norm": 6.5620012283325195,
       "learning_rate": 3.0214424951267055e-05,
-      "loss": 0.2054,
+      "loss": 0.2285,
       "step": 260
     },
     {
       "epoch": 1.4210526315789473,
-      "grad_norm": 6.9045610427856445,
+      "grad_norm": 12.308910369873047,
       "learning_rate": 2.9239766081871346e-05,
-      "loss": 0.217,
+      "loss": 0.1919,
       "step": 270
     },
     {
       "epoch": 1.4736842105263157,
-      "grad_norm": 8.878619194030762,
+      "grad_norm": 11.529248237609863,
       "learning_rate": 2.8265107212475634e-05,
-      "loss": 0.2074,
+      "loss": 0.183,
       "step": 280
     },
     {
       "epoch": 1.526315789473684,
-      "grad_norm": 9.569494247436523,
+      "grad_norm": 7.2859930992126465,
       "learning_rate": 2.729044834307992e-05,
-      "loss": 0.2046,
+      "loss": 0.1908,
       "step": 290
     },
     {
       "epoch": 1.5789473684210527,
-      "grad_norm": 10.316445350646973,
+      "grad_norm": 6.837076187133789,
       "learning_rate": 2.6315789473684212e-05,
-      "loss": 0.1704,
+      "loss": 0.1915,
       "step": 300
     },
     {
       "epoch": 1.631578947368421,
-      "grad_norm": 14.888699531555176,
+      "grad_norm": 7.783897876739502,
       "learning_rate": 2.53411306042885e-05,
-      "loss": 0.1674,
+      "loss": 0.1756,
       "step": 310
     },
     {
       "epoch": 1.6842105263157894,
-      "grad_norm": 11.373708724975586,
+      "grad_norm": 10.164815902709961,
       "learning_rate": 2.4366471734892787e-05,
-      "loss": 0.1637,
+      "loss": 0.1832,
       "step": 320
     },
     {
       "epoch": 1.736842105263158,
-      "grad_norm": 13.694819450378418,
+      "grad_norm": 8.077231407165527,
       "learning_rate": 2.3391812865497074e-05,
       "loss": 0.1904,
       "step": 330
     },
     {
       "epoch": 1.7894736842105263,
-      "grad_norm": 11.21845817565918,
+      "grad_norm": 7.405230522155762,
       "learning_rate": 2.2417153996101365e-05,
-      "loss": 0.1539,
+      "loss": 0.198,
       "step": 340
     },
     {
       "epoch": 1.8421052631578947,
-      "grad_norm": 9.454412460327148,
+      "grad_norm": 15.994668006896973,
       "learning_rate": 2.1442495126705653e-05,
-      "loss": 0.1953,
+      "loss": 0.1899,
       "step": 350
     },
     {
       "epoch": 1.8947368421052633,
-      "grad_norm": 11.822175979614258,
+      "grad_norm": 9.648083686828613,
       "learning_rate": 2.046783625730994e-05,
-      "loss": 0.1608,
+      "loss": 0.2015,
       "step": 360
     },
     {
       "epoch": 1.9473684210526314,
-      "grad_norm": 18.53520393371582,
+      "grad_norm": 10.014592170715332,
       "learning_rate": 1.949317738791423e-05,
-      "loss": 0.1849,
+      "loss": 0.1738,
       "step": 370
     },
     {
       "epoch": 2.0,
-      "grad_norm": 9.514893531799316,
+      "grad_norm": 13.817649841308594,
       "learning_rate": 1.8518518518518518e-05,
-      "loss": 0.1862,
+      "loss": 0.1963,
       "step": 380
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.9748148148148148,
-      "eval_loss": 0.07534925639629364,
-      "eval_runtime": 16.2086,
-      "eval_samples_per_second": 166.578,
-      "eval_steps_per_second": 5.244,
+      "eval_accuracy": 0.9714814814814815,
+      "eval_loss": 0.09753727912902832,
+      "eval_runtime": 15.9787,
+      "eval_samples_per_second": 168.975,
+      "eval_steps_per_second": 5.32,
       "step": 380
     },
     {
       "epoch": 2.0526315789473686,
-      "grad_norm": 7.307323932647705,
+      "grad_norm": 9.863332748413086,
       "learning_rate": 1.7543859649122806e-05,
-      "loss": 0.1862,
+      "loss": 0.1578,
       "step": 390
     },
     {
       "epoch": 2.1052631578947367,
-      "grad_norm": 12.584000587463379,
+      "grad_norm": 5.009075164794922,
       "learning_rate": 1.6569200779727097e-05,
-      "loss": 0.1865,
+      "loss": 0.1485,
       "step": 400
     },
     {
       "epoch": 2.1578947368421053,
-      "grad_norm": 11.615283966064453,
+      "grad_norm": 7.887981414794922,
       "learning_rate": 1.5594541910331384e-05,
-      "loss": 0.1706,
+      "loss": 0.1562,
       "step": 410
     },
     {
       "epoch": 2.2105263157894735,
-      "grad_norm": 14.084049224853516,
+      "grad_norm": 8.103832244873047,
       "learning_rate": 1.4619883040935673e-05,
-      "loss": 0.1775,
+      "loss": 0.1634,
       "step": 420
     },
     {
       "epoch": 2.263157894736842,
-      "grad_norm": 7.380066394805908,
+      "grad_norm": 10.090664863586426,
       "learning_rate": 1.364522417153996e-05,
-      "loss": 0.1578,
+      "loss": 0.152,
       "step": 430
     },
     {
       "epoch": 2.3157894736842106,
-      "grad_norm": 5.233687877655029,
+      "grad_norm": 4.781908988952637,
       "learning_rate": 1.267056530214425e-05,
-      "loss": 0.1548,
+      "loss": 0.1575,
       "step": 440
     },
     {
       "epoch": 2.3684210526315788,
-      "grad_norm": 6.7702765464782715,
+      "grad_norm": 6.325347900390625,
       "learning_rate": 1.1695906432748537e-05,
-      "loss": 0.1618,
+      "loss": 0.1435,
       "step": 450
     },
     {
       "epoch": 2.4210526315789473,
-      "grad_norm": 10.425500869750977,
+      "grad_norm": 9.535646438598633,
       "learning_rate": 1.0721247563352826e-05,
-      "loss": 0.1174,
+      "loss": 0.1302,
       "step": 460
     },
     {
       "epoch": 2.473684210526316,
-      "grad_norm": 9.966532707214355,
+      "grad_norm": 16.191673278808594,
       "learning_rate": 9.746588693957115e-06,
-      "loss": 0.1395,
+      "loss": 0.1455,
       "step": 470
     },
     {
       "epoch": 2.526315789473684,
-      "grad_norm": 8.691372871398926,
+      "grad_norm": 6.5809783935546875,
       "learning_rate": 8.771929824561403e-06,
-      "loss": 0.168,
+      "loss": 0.1496,
       "step": 480
     },
     {
       "epoch": 2.5789473684210527,
-      "grad_norm": 8.912318229675293,
+      "grad_norm": 6.148409843444824,
       "learning_rate": 7.797270955165692e-06,
-      "loss": 0.1366,
+      "loss": 0.14,
       "step": 490
     },
     {
       "epoch": 2.6315789473684212,
-      "grad_norm": 8.191598892211914,
+      "grad_norm": 9.369363784790039,
       "learning_rate": 6.82261208576998e-06,
-      "loss": 0.128,
+      "loss": 0.1535,
       "step": 500
     },
     {
       "epoch": 2.6842105263157894,
-      "grad_norm": 11.146318435668945,
+      "grad_norm": 5.0577521324157715,
       "learning_rate": 5.8479532163742686e-06,
-      "loss": 0.1428,
+      "loss": 0.1471,
       "step": 510
     },
     {
       "epoch": 2.736842105263158,
-      "grad_norm": 9.236715316772461,
+      "grad_norm": 5.934864044189453,
       "learning_rate": 4.873294346978558e-06,
-      "loss": 0.1347,
+      "loss": 0.1429,
       "step": 520
     },
     {
       "epoch": 2.7894736842105265,
-      "grad_norm": 6.167862892150879,
+      "grad_norm": 5.143869400024414,
       "learning_rate": 3.898635477582846e-06,
-      "loss": 0.172,
+      "loss": 0.1278,
       "step": 530
     },
     {
       "epoch": 2.8421052631578947,
-      "grad_norm": 11.211396217346191,
+      "grad_norm": 12.090688705444336,
       "learning_rate": 2.9239766081871343e-06,
-      "loss": 0.162,
+      "loss": 0.141,
       "step": 540
     },
     {
       "epoch": 2.8947368421052633,
-      "grad_norm": 12.20263385772705,
+      "grad_norm": 8.571892738342285,
       "learning_rate": 1.949317738791423e-06,
-      "loss": 0.1697,
+      "loss": 0.1296,
       "step": 550
     },
     {
       "epoch": 2.9473684210526314,
-      "grad_norm": 6.672098159790039,
+      "grad_norm": 14.7016019821167,
       "learning_rate": 9.746588693957115e-07,
-      "loss": 0.1561,
+      "loss": 0.1363,
       "step": 560
     },
     {
       "epoch": 3.0,
-      "grad_norm": 11.088051795959473,
+      "grad_norm": 4.905235767364502,
       "learning_rate": 0.0,
-      "loss": 0.1673,
+      "loss": 0.1511,
       "step": 570
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.9781481481481481,
-      "eval_loss": 0.062276002019643784,
-      "eval_runtime": 16.8944,
-      "eval_samples_per_second": 159.816,
-      "eval_steps_per_second": 5.031,
+      "eval_accuracy": 0.9740740740740741,
+      "eval_loss": 0.07933878898620605,
+      "eval_runtime": 16.0704,
+      "eval_samples_per_second": 168.011,
+      "eval_steps_per_second": 5.289,
       "step": 570
     },
     {
       "epoch": 3.0,
       "step": 570,
       "total_flos": 1.8124066505760768e+18,
-      "train_loss": 0.3407971503441794,
-      "train_runtime": 1119.3728,
-      "train_samples_per_second": 65.126,
-      "train_steps_per_second": 0.509
+      "train_loss": 0.32989514710610374,
+      "train_runtime": 1304.6847,
+      "train_samples_per_second": 55.876,
+      "train_steps_per_second": 0.437
     }
   ],
   "logging_steps": 10,