Training in progress, epoch 1

all_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 3.0,
     "total_flos": 8.923548605812163e+17,
-    "train_loss": 0.17324901176965174,
-    "train_runtime": 1100.5808,
-    "train_samples_per_second": 32.62,
-    "train_steps_per_second": 1.019
+    "train_loss": 0.0954810879500367,
+    "train_runtime": 1092.3028,
+    "train_samples_per_second": 32.867,
+    "train_steps_per_second": 1.027
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:507c54bd18248eb46ee8424472d91bd83fa78ea044b05ca56e8b7da55f23270d
+oid sha256:3f561c57bb607b56d41e861c33cec85216d889d6ddefe725c73e7810f9c537fa
 size 110342832

runs/Dec07_11-35-50_9b6bffedd0df/events.out.tfevents.1701951313.9b6bffedd0df.300.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:feb8599f35f8e1879c71ebde219d9c3eda514ab38c20b72e8cc760439eaa9a9c
+size 10845

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 3.0,
     "total_flos": 8.923548605812163e+17,
-    "train_loss": 0.17324901176965174,
-    "train_runtime": 1100.5808,
-    "train_samples_per_second": 32.62,
-    "train_steps_per_second": 1.019
+    "train_loss": 0.0954810879500367,
+    "train_runtime": 1092.3028,
+    "train_samples_per_second": 32.867,
+    "train_steps_per_second": 1.027
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.9856283422459893,
-  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-748",
+  "best_metric": 0.9832887700534759,
+  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-374",
   "epoch": 3.0,
   "eval_steps": 500,
   "global_step": 1122,
@@ -11,710 +11,710 @@
     {
       "epoch": 0.03,
       "learning_rate": 4.424778761061947e-06,
-      "loss": 0.7462,
+      "loss": 0.0941,
       "step": 10
     },
     {
       "epoch": 0.05,
       "learning_rate": 8.849557522123894e-06,
-      "loss": 0.6988,
+      "loss": 0.1177,
       "step": 20
     },
     {
       "epoch": 0.08,
       "learning_rate": 1.3274336283185843e-05,
-      "loss": 0.6181,
+      "loss": 0.0884,
       "step": 30
     },
     {
       "epoch": 0.11,
       "learning_rate": 1.7699115044247787e-05,
-      "loss": 0.5062,
+      "loss": 0.0714,
       "step": 40
     },
     {
       "epoch": 0.13,
       "learning_rate": 2.2123893805309738e-05,
-      "loss": 0.4631,
+      "loss": 0.0417,
       "step": 50
     },
     {
       "epoch": 0.16,
       "learning_rate": 2.6548672566371686e-05,
-      "loss": 0.4307,
+      "loss": 0.2017,
       "step": 60
     },
     {
       "epoch": 0.19,
       "learning_rate": 3.097345132743363e-05,
-      "loss": 0.401,
+      "loss": 0.1243,
       "step": 70
     },
     {
       "epoch": 0.21,
       "learning_rate": 3.5398230088495574e-05,
-      "loss": 0.3885,
+      "loss": 0.1474,
       "step": 80
     },
     {
       "epoch": 0.24,
       "learning_rate": 3.982300884955752e-05,
-      "loss": 0.3324,
+      "loss": 0.0888,
       "step": 90
     },
     {
       "epoch": 0.27,
       "learning_rate": 4.4247787610619477e-05,
-      "loss": 0.3699,
+      "loss": 0.1036,
       "step": 100
     },
     {
       "epoch": 0.29,
       "learning_rate": 4.867256637168142e-05,
-      "loss": 0.3131,
+      "loss": 0.1529,
       "step": 110
     },
     {
       "epoch": 0.32,
       "learning_rate": 4.9653121902874134e-05,
-      "loss": 0.3063,
+      "loss": 0.0893,
       "step": 120
     },
     {
       "epoch": 0.35,
       "learning_rate": 4.9157581764122896e-05,
-      "loss": 0.2404,
+      "loss": 0.1547,
       "step": 130
     },
     {
       "epoch": 0.37,
       "learning_rate": 4.866204162537166e-05,
-      "loss": 0.2354,
+      "loss": 0.1489,
       "step": 140
     },
     {
       "epoch": 0.4,
       "learning_rate": 4.816650148662042e-05,
-      "loss": 0.3317,
+      "loss": 0.1667,
       "step": 150
     },
     {
       "epoch": 0.43,
       "learning_rate": 4.767096134786918e-05,
-      "loss": 0.2295,
+      "loss": 0.089,
       "step": 160
     },
     {
       "epoch": 0.45,
       "learning_rate": 4.717542120911794e-05,
-      "loss": 0.2367,
+      "loss": 0.1242,
       "step": 170
     },
     {
       "epoch": 0.48,
       "learning_rate": 4.66798810703667e-05,
-      "loss": 0.2368,
+      "loss": 0.134,
       "step": 180
     },
     {
       "epoch": 0.51,
       "learning_rate": 4.618434093161546e-05,
-      "loss": 0.2302,
+      "loss": 0.1412,
       "step": 190
     },
     {
       "epoch": 0.53,
       "learning_rate": 4.568880079286422e-05,
-      "loss": 0.2478,
+      "loss": 0.106,
       "step": 200
     },
     {
       "epoch": 0.56,
       "learning_rate": 4.5193260654112984e-05,
-      "loss": 0.253,
+      "loss": 0.142,
       "step": 210
     },
     {
       "epoch": 0.59,
       "learning_rate": 4.4697720515361746e-05,
-      "loss": 0.2002,
+      "loss": 0.074,
       "step": 220
     },
     {
       "epoch": 0.61,
       "learning_rate": 4.420218037661051e-05,
-      "loss": 0.2029,
+      "loss": 0.1302,
       "step": 230
     },
     {
       "epoch": 0.64,
       "learning_rate": 4.370664023785927e-05,
-      "loss": 0.2061,
+      "loss": 0.1543,
       "step": 240
     },
     {
       "epoch": 0.67,
       "learning_rate": 4.321110009910803e-05,
-      "loss": 0.2152,
+      "loss": 0.1548,
       "step": 250
     },
     {
       "epoch": 0.7,
       "learning_rate": 4.27155599603568e-05,
-      "loss": 0.2375,
+      "loss": 0.1783,
       "step": 260
     },
     {
       "epoch": 0.72,
       "learning_rate": 4.222001982160555e-05,
-      "loss": 0.1803,
+      "loss": 0.1362,
       "step": 270
     },
     {
       "epoch": 0.75,
       "learning_rate": 4.172447968285431e-05,
-      "loss": 0.2022,
+      "loss": 0.094,
       "step": 280
     },
     {
       "epoch": 0.78,
       "learning_rate": 4.122893954410307e-05,
-      "loss": 0.2074,
+      "loss": 0.0971,
       "step": 290
     },
     {
       "epoch": 0.8,
       "learning_rate": 4.0733399405351834e-05,
-      "loss": 0.1728,
+      "loss": 0.1013,
       "step": 300
     },
     {
       "epoch": 0.83,
       "learning_rate": 4.0237859266600596e-05,
-      "loss": 0.209,
+      "loss": 0.1397,
       "step": 310
     },
     {
       "epoch": 0.86,
       "learning_rate": 3.9742319127849357e-05,
-      "loss": 0.1783,
+      "loss": 0.1397,
       "step": 320
     },
     {
       "epoch": 0.88,
       "learning_rate": 3.924677898909812e-05,
-      "loss": 0.1595,
+      "loss": 0.1575,
       "step": 330
     },
     {
       "epoch": 0.91,
       "learning_rate": 3.8751238850346886e-05,
-      "loss": 0.2652,
+      "loss": 0.1597,
       "step": 340
     },
     {
       "epoch": 0.94,
       "learning_rate": 3.825569871159564e-05,
-      "loss": 0.1958,
+      "loss": 0.0972,
       "step": 350
     },
     {
       "epoch": 0.96,
       "learning_rate": 3.77601585728444e-05,
-      "loss": 0.1557,
+      "loss": 0.1529,
       "step": 360
     },
     {
       "epoch": 0.99,
       "learning_rate": 3.726461843409316e-05,
-      "loss": 0.1314,
+      "loss": 0.0816,
       "step": 370
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.9705882352941176,
-      "eval_loss": 0.07545506209135056,
-      "eval_runtime": 53.5066,
-      "eval_samples_per_second": 55.918,
-      "eval_steps_per_second": 6.99,
+      "eval_accuracy": 0.9832887700534759,
+      "eval_loss": 0.04817664995789528,
+      "eval_runtime": 56.5513,
+      "eval_samples_per_second": 52.908,
+      "eval_steps_per_second": 6.613,
       "step": 374
     },
     {
       "epoch": 1.02,
       "learning_rate": 3.676907829534192e-05,
-      "loss": 0.1853,
+      "loss": 0.1262,
       "step": 380
     },
     {
       "epoch": 1.04,
       "learning_rate": 3.6273538156590684e-05,
-      "loss": 0.1775,
+      "loss": 0.106,
       "step": 390
     },
     {
       "epoch": 1.07,
       "learning_rate": 3.5777998017839445e-05,
-      "loss": 0.1357,
+      "loss": 0.0638,
       "step": 400
     },
     {
       "epoch": 1.1,
       "learning_rate": 3.5282457879088207e-05,
-      "loss": 0.1917,
+      "loss": 0.1009,
       "step": 410
     },
     {
       "epoch": 1.12,
       "learning_rate": 3.4786917740336974e-05,
-      "loss": 0.123,
+      "loss": 0.0882,
       "step": 420
     },
     {
       "epoch": 1.15,
       "learning_rate": 3.429137760158573e-05,
-      "loss": 0.1059,
+      "loss": 0.1135,
       "step": 430
     },
     {
       "epoch": 1.18,
       "learning_rate": 3.379583746283449e-05,
-      "loss": 0.1621,
+      "loss": 0.0714,
       "step": 440
     },
     {
       "epoch": 1.2,
       "learning_rate": 3.330029732408325e-05,
-      "loss": 0.1796,
+      "loss": 0.1241,
       "step": 450
     },
     {
       "epoch": 1.23,
       "learning_rate": 3.280475718533201e-05,
-      "loss": 0.1419,
+      "loss": 0.1248,
       "step": 460
     },
     {
       "epoch": 1.26,
       "learning_rate": 3.230921704658077e-05,
-      "loss": 0.1244,
+      "loss": 0.1366,
       "step": 470
     },
     {
       "epoch": 1.28,
       "learning_rate": 3.1813676907829534e-05,
-      "loss": 0.1493,
+      "loss": 0.0786,
       "step": 480
     },
     {
       "epoch": 1.31,
       "learning_rate": 3.13181367690783e-05,
-      "loss": 0.1554,
+      "loss": 0.1085,
       "step": 490
     },
     {
       "epoch": 1.34,
       "learning_rate": 3.0822596630327057e-05,
-      "loss": 0.1269,
+      "loss": 0.0979,
       "step": 500
     },
     {
       "epoch": 1.36,
       "learning_rate": 3.0327056491575818e-05,
-      "loss": 0.1522,
+      "loss": 0.0745,
       "step": 510
     },
     {
       "epoch": 1.39,
       "learning_rate": 2.983151635282458e-05,
-      "loss": 0.1354,
+      "loss": 0.0741,
       "step": 520
     },
     {
       "epoch": 1.42,
       "learning_rate": 2.933597621407334e-05,
-      "loss": 0.1673,
+      "loss": 0.1569,
       "step": 530
     },
     {
       "epoch": 1.44,
       "learning_rate": 2.88404360753221e-05,
-      "loss": 0.1184,
+      "loss": 0.1531,
       "step": 540
     },
     {
       "epoch": 1.47,
       "learning_rate": 2.8344895936570865e-05,
-      "loss": 0.1498,
+      "loss": 0.086,
       "step": 550
     },
     {
       "epoch": 1.5,
       "learning_rate": 2.7849355797819627e-05,
-      "loss": 0.1333,
+      "loss": 0.1204,
       "step": 560
     },
     {
       "epoch": 1.52,
       "learning_rate": 2.7353815659068388e-05,
-      "loss": 0.1146,
+      "loss": 0.0477,
       "step": 570
     },
     {
       "epoch": 1.55,
       "learning_rate": 2.6858275520317145e-05,
-      "loss": 0.1239,
+      "loss": 0.0972,
       "step": 580
     },
     {
       "epoch": 1.58,
       "learning_rate": 2.6362735381565906e-05,
-      "loss": 0.1137,
+      "loss": 0.0847,
       "step": 590
     },
     {
       "epoch": 1.6,
       "learning_rate": 2.5867195242814668e-05,
-      "loss": 0.087,
+      "loss": 0.1807,
       "step": 600
     },
     {
       "epoch": 1.63,
       "learning_rate": 2.537165510406343e-05,
-      "loss": 0.1045,
+      "loss": 0.0957,
       "step": 610
     },
     {
       "epoch": 1.66,
       "learning_rate": 2.4876114965312193e-05,
-      "loss": 0.1408,
+      "loss": 0.053,
       "step": 620
     },
     {
       "epoch": 1.68,
       "learning_rate": 2.4380574826560954e-05,
-      "loss": 0.1048,
+      "loss": 0.081,
       "step": 630
     },
     {
       "epoch": 1.71,
       "learning_rate": 2.3885034687809712e-05,
-      "loss": 0.1363,
+      "loss": 0.081,
       "step": 640
     },
     {
       "epoch": 1.74,
       "learning_rate": 2.3389494549058473e-05,
-      "loss": 0.1342,
+      "loss": 0.0824,
       "step": 650
     },
     {
       "epoch": 1.76,
       "learning_rate": 2.2893954410307238e-05,
-      "loss": 0.0823,
+      "loss": 0.0451,
       "step": 660
     },
     {
       "epoch": 1.79,
       "learning_rate": 2.2398414271556e-05,
-      "loss": 0.1348,
+      "loss": 0.0624,
       "step": 670
     },
     {
       "epoch": 1.82,
       "learning_rate": 2.1902874132804756e-05,
-      "loss": 0.127,
+      "loss": 0.082,
       "step": 680
     },
     {
       "epoch": 1.84,
       "learning_rate": 2.1407333994053518e-05,
-      "loss": 0.1397,
+      "loss": 0.0664,
       "step": 690
     },
     {
       "epoch": 1.87,
       "learning_rate": 2.0911793855302282e-05,
-      "loss": 0.1235,
+      "loss": 0.1328,
       "step": 700
     },
     {
       "epoch": 1.9,
       "learning_rate": 2.0416253716551043e-05,
-      "loss": 0.085,
+      "loss": 0.1391,
       "step": 710
     },
     {
       "epoch": 1.93,
       "learning_rate": 1.99207135777998e-05,
-      "loss": 0.1204,
+      "loss": 0.1112,
       "step": 720
     },
     {
       "epoch": 1.95,
       "learning_rate": 1.9425173439048562e-05,
-      "loss": 0.1456,
+      "loss": 0.06,
       "step": 730
     },
     {
       "epoch": 1.98,
       "learning_rate": 1.8929633300297326e-05,
-      "loss": 0.1,
+      "loss": 0.0598,
       "step": 740
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.9856283422459893,
-      "eval_loss": 0.04481882601976395,
-      "eval_runtime": 50.8588,
-      "eval_samples_per_second": 58.83,
-      "eval_steps_per_second": 7.354,
+      "eval_accuracy": 0.9802807486631016,
+      "eval_loss": 0.06319520622491837,
+      "eval_runtime": 55.8699,
+      "eval_samples_per_second": 53.553,
+      "eval_steps_per_second": 6.694,
       "step": 748
     },
     {
       "epoch": 2.01,
       "learning_rate": 1.8434093161546088e-05,
-      "loss": 0.0959,
+      "loss": 0.0793,
       "step": 750
     },
     {
       "epoch": 2.03,
       "learning_rate": 1.7938553022794845e-05,
-      "loss": 0.1193,
+      "loss": 0.0988,
       "step": 760
     },
     {
       "epoch": 2.06,
       "learning_rate": 1.7443012884043606e-05,
-      "loss": 0.0731,
+      "loss": 0.0534,
       "step": 770
     },
     {
       "epoch": 2.09,
       "learning_rate": 1.694747274529237e-05,
-      "loss": 0.1176,
+      "loss": 0.0979,
       "step": 780
     },
     {
       "epoch": 2.11,
       "learning_rate": 1.6451932606541132e-05,
-      "loss": 0.0947,
+      "loss": 0.0652,
       "step": 790
     },
     {
       "epoch": 2.14,
       "learning_rate": 1.595639246778989e-05,
-      "loss": 0.1183,
+      "loss": 0.0655,
       "step": 800
     },
     {
       "epoch": 2.17,
       "learning_rate": 1.5460852329038654e-05,
-      "loss": 0.1378,
+      "loss": 0.0513,
       "step": 810
     },
     {
       "epoch": 2.19,
       "learning_rate": 1.4965312190287414e-05,
-      "loss": 0.0664,
+      "loss": 0.0964,
       "step": 820
     },
     {
       "epoch": 2.22,
       "learning_rate": 1.4469772051536176e-05,
-      "loss": 0.0801,
+      "loss": 0.0828,
       "step": 830
     },
     {
       "epoch": 2.25,
       "learning_rate": 1.3974231912784936e-05,
-      "loss": 0.1572,
+      "loss": 0.0913,
       "step": 840
     },
     {
       "epoch": 2.27,
       "learning_rate": 1.3478691774033697e-05,
-      "loss": 0.1291,
+      "loss": 0.0488,
       "step": 850
     },
     {
       "epoch": 2.3,
       "learning_rate": 1.298315163528246e-05,
-      "loss": 0.1124,
+      "loss": 0.1146,
       "step": 860
     },
     {
       "epoch": 2.33,
       "learning_rate": 1.2487611496531219e-05,
-      "loss": 0.1049,
+      "loss": 0.0858,
       "step": 870
     },
     {
       "epoch": 2.35,
       "learning_rate": 1.1992071357779982e-05,
-      "loss": 0.1002,
+      "loss": 0.0955,
       "step": 880
     },
     {
       "epoch": 2.38,
       "learning_rate": 1.1496531219028741e-05,
-      "loss": 0.0843,
+      "loss": 0.03,
       "step": 890
     },
     {
       "epoch": 2.41,
       "learning_rate": 1.1000991080277504e-05,
-      "loss": 0.1149,
+      "loss": 0.0425,
       "step": 900
     },
     {
       "epoch": 2.43,
       "learning_rate": 1.0505450941526264e-05,
-      "loss": 0.1035,
+      "loss": 0.083,
       "step": 910
     },
     {
       "epoch": 2.46,
       "learning_rate": 1.0009910802775026e-05,
-      "loss": 0.0833,
+      "loss": 0.0778,
       "step": 920
     },
     {
       "epoch": 2.49,
       "learning_rate": 9.514370664023786e-06,
-      "loss": 0.1189,
+      "loss": 0.0936,
       "step": 930
     },
     {
       "epoch": 2.51,
       "learning_rate": 9.018830525272549e-06,
-      "loss": 0.0674,
+      "loss": 0.0553,
       "step": 940
     },
     {
       "epoch": 2.54,
       "learning_rate": 8.523290386521308e-06,
-      "loss": 0.071,
+      "loss": 0.0378,
       "step": 950
     },
     {
       "epoch": 2.57,
       "learning_rate": 8.02775024777007e-06,
-      "loss": 0.052,
+      "loss": 0.0444,
       "step": 960
     },
     {
       "epoch": 2.59,
       "learning_rate": 7.532210109018831e-06,
-      "loss": 0.0916,
+      "loss": 0.0487,
       "step": 970
     },
     {
       "epoch": 2.62,
       "learning_rate": 7.036669970267593e-06,
-      "loss": 0.1102,
+      "loss": 0.0609,
       "step": 980
     },
     {
       "epoch": 2.65,
       "learning_rate": 6.541129831516353e-06,
-      "loss": 0.091,
+      "loss": 0.0812,
       "step": 990
     },
     {
       "epoch": 2.67,
       "learning_rate": 6.045589692765114e-06,
-      "loss": 0.1315,
+      "loss": 0.0952,
       "step": 1000
     },
     {
       "epoch": 2.7,
       "learning_rate": 5.5500495540138754e-06,
-      "loss": 0.0655,
+      "loss": 0.0673,
       "step": 1010
     },
     {
       "epoch": 2.73,
       "learning_rate": 5.0545094152626366e-06,
-      "loss": 0.0558,
+      "loss": 0.045,
       "step": 1020
     },
     {
       "epoch": 2.75,
       "learning_rate": 4.558969276511398e-06,
-      "loss": 0.0807,
+      "loss": 0.0725,
       "step": 1030
     },
     {
       "epoch": 2.78,
       "learning_rate": 4.063429137760159e-06,
-      "loss": 0.0695,
+      "loss": 0.0835,
       "step": 1040
     },
     {
       "epoch": 2.81,
       "learning_rate": 3.56788899900892e-06,
-      "loss": 0.0731,
+      "loss": 0.0773,
       "step": 1050
     },
     {
       "epoch": 2.83,
       "learning_rate": 3.072348860257681e-06,
-      "loss": 0.0919,
+      "loss": 0.0783,
       "step": 1060
     },
     {
       "epoch": 2.86,
       "learning_rate": 2.576808721506442e-06,
-      "loss": 0.0611,
+      "loss": 0.0518,
       "step": 1070
     },
     {
       "epoch": 2.89,
       "learning_rate": 2.081268582755203e-06,
-      "loss": 0.0587,
+      "loss": 0.013,
       "step": 1080
     },
     {
       "epoch": 2.91,
       "learning_rate": 1.5857284440039643e-06,
-      "loss": 0.0965,
+      "loss": 0.0471,
       "step": 1090
     },
     {
       "epoch": 2.94,
       "learning_rate": 1.0901883052527254e-06,
-      "loss": 0.0988,
+      "loss": 0.043,
       "step": 1100
     },
     {
       "epoch": 2.97,
       "learning_rate": 5.946481665014866e-07,
-      "loss": 0.0697,
+      "loss": 0.0377,
       "step": 1110
     },
     {
       "epoch": 2.99,
       "learning_rate": 9.910802775024777e-08,
-      "loss": 0.0919,
+      "loss": 0.0581,
       "step": 1120
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.9836229946524064,
-      "eval_loss": 0.042694564908742905,
-      "eval_runtime": 51.0671,
-      "eval_samples_per_second": 58.59,
-      "eval_steps_per_second": 7.324,
+      "eval_accuracy": 0.982620320855615,
+      "eval_loss": 0.0579056441783905,
+      "eval_runtime": 58.7043,
+      "eval_samples_per_second": 50.967,
+      "eval_steps_per_second": 6.371,
       "step": 1122
     },
     {
       "epoch": 3.0,
       "step": 1122,
       "total_flos": 8.923548605812163e+17,
-      "train_loss": 0.17324901176965174,
-      "train_runtime": 1100.5808,
-      "train_samples_per_second": 32.62,
-      "train_steps_per_second": 1.019
+      "train_loss": 0.0954810879500367,
+      "train_runtime": 1092.3028,
+      "train_samples_per_second": 32.867,
+      "train_steps_per_second": 1.027
     }
   ],
   "logging_steps": 10,