Training in progress, epoch 1

all_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 10.0,
     "total_flos": 6.954705718242509e+17,
-    "train_loss": 0.2417184677991,
-    "train_runtime": 1633.5967,
-    "train_samples_per_second": 17.128,
-    "train_steps_per_second": 0.135
+    "train_loss": 0.23402113264257257,
+    "train_runtime": 1621.2514,
+    "train_samples_per_second": 17.258,
+    "train_steps_per_second": 0.136
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5fd62ab7f09533aa4d4bbb34db59c9314e218eae205a9fcdf48f1e6408a8f8cc
+oid sha256:6e54bdc6cac7d51387317dc47d851fc2b73485816c5eb83c4e7c48ce7968c813
 size 110342832

runs/Oct26_08-08-51_c3ee2852f8cc/events.out.tfevents.1729936292.c3ee2852f8cc.307.3

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d7cd695ec956a2a43ae758fbf1280c6c5a788bdb5d19455dc06e54d678dd3b99
+size 6323

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 10.0,
     "total_flos": 6.954705718242509e+17,
-    "train_loss": 0.2417184677991,
-    "train_runtime": 1633.5967,
-    "train_samples_per_second": 17.128,
-    "train_steps_per_second": 0.135
+    "train_loss": 0.23402113264257257,
+    "train_runtime": 1621.2514,
+    "train_samples_per_second": 17.258,
+    "train_steps_per_second": 0.136
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.8873642285098018,
-  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-132",
+  "best_metric": 0.8866264991906878,
+  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-220",
   "epoch": 10.0,
   "eval_steps": 500,
   "global_step": 220,
@@ -10,286 +10,286 @@
   "log_history": [
     {
       "epoch": 0.45454545454545453,
-      "grad_norm": 25.305431365966797,
+      "grad_norm": 3.2964470386505127,
       "learning_rate": 2.272727272727273e-05,
-      "loss": 0.2546,
+      "loss": 0.2322,
       "step": 10
     },
     {
       "epoch": 0.9090909090909091,
-      "grad_norm": 22.30314826965332,
+      "grad_norm": 2.9043655395507812,
       "learning_rate": 4.545454545454546e-05,
-      "loss": 0.2646,
+      "loss": 0.2604,
       "step": 20
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.7556270096463023,
-      "eval_f1": 0.6993854476093359,
-      "eval_loss": 0.7455090880393982,
-      "eval_precision": 0.8203390210291629,
-      "eval_recall": 0.7556270096463023,
-      "eval_runtime": 15.9447,
-      "eval_samples_per_second": 19.505,
-      "eval_steps_per_second": 0.627,
+      "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,
       "step": 22
     },
     {
       "epoch": 1.3636363636363638,
-      "grad_norm": 12.352283477783203,
+      "grad_norm": 56.99660873413086,
       "learning_rate": 4.797979797979798e-05,
-      "loss": 0.2258,
+      "loss": 0.2223,
       "step": 30
     },
     {
       "epoch": 1.8181818181818183,
-      "grad_norm": 16.673419952392578,
+      "grad_norm": 18.48450469970703,
       "learning_rate": 4.545454545454546e-05,
-      "loss": 0.2359,
+      "loss": 0.2492,
       "step": 40
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8102893890675241,
-      "eval_f1": 0.8028661921793899,
-      "eval_loss": 0.29898953437805176,
-      "eval_precision": 0.8059591355411291,
-      "eval_recall": 0.8102893890675241,
-      "eval_runtime": 16.396,
-      "eval_samples_per_second": 18.968,
-      "eval_steps_per_second": 0.61,
+      "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,
       "step": 44
     },
     {
       "epoch": 2.2727272727272725,
-      "grad_norm": 166.52926635742188,
+      "grad_norm": 6.249089241027832,
       "learning_rate": 4.292929292929293e-05,
-      "loss": 0.2831,
+      "loss": 0.2374,
       "step": 50
     },
     {
       "epoch": 2.7272727272727275,
-      "grad_norm": 6.889273643493652,
+      "grad_norm": 34.324771881103516,
       "learning_rate": 4.0404040404040405e-05,
-      "loss": 0.2477,
+      "loss": 0.2373,
       "step": 60
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.8263665594855305,
-      "eval_f1": 0.8089675597714183,
-      "eval_loss": 0.2808948755264282,
-      "eval_precision": 0.8441865239067687,
-      "eval_recall": 0.8263665594855305,
-      "eval_runtime": 17.6236,
-      "eval_samples_per_second": 17.647,
-      "eval_steps_per_second": 0.567,
+      "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,
       "step": 66
     },
     {
       "epoch": 3.1818181818181817,
-      "grad_norm": 5.792304039001465,
+      "grad_norm": 12.280445098876953,
       "learning_rate": 3.787878787878788e-05,
-      "loss": 0.2553,
+      "loss": 0.2634,
       "step": 70
     },
     {
       "epoch": 3.6363636363636362,
-      "grad_norm": 33.26253128051758,
+      "grad_norm": 4.722353935241699,
       "learning_rate": 3.535353535353535e-05,
-      "loss": 0.2476,
+      "loss": 0.2623,
       "step": 80
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.8327974276527331,
-      "eval_f1": 0.8308694134948529,
-      "eval_loss": 0.22933611273765564,
-      "eval_precision": 0.8301699258286112,
-      "eval_recall": 0.8327974276527331,
-      "eval_runtime": 15.7477,
-      "eval_samples_per_second": 19.749,
-      "eval_steps_per_second": 0.635,
+      "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,
       "step": 88
     },
     {
       "epoch": 4.090909090909091,
-      "grad_norm": 36.04878234863281,
+      "grad_norm": 11.214787483215332,
       "learning_rate": 3.282828282828283e-05,
-      "loss": 0.2594,
+      "loss": 0.2246,
       "step": 90
     },
     {
       "epoch": 4.545454545454545,
-      "grad_norm": 12.086928367614746,
+      "grad_norm": 13.322309494018555,
       "learning_rate": 3.0303030303030306e-05,
-      "loss": 0.251,
+      "loss": 0.232,
       "step": 100
     },
     {
       "epoch": 5.0,
-      "grad_norm": 7.808885097503662,
+      "grad_norm": 8.625347137451172,
       "learning_rate": 2.777777777777778e-05,
-      "loss": 0.2676,
+      "loss": 0.2378,
       "step": 110
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.8231511254019293,
-      "eval_f1": 0.8202510778719536,
-      "eval_loss": 0.25844427943229675,
-      "eval_precision": 0.8196657042171692,
-      "eval_recall": 0.8231511254019293,
-      "eval_runtime": 15.8923,
-      "eval_samples_per_second": 19.569,
-      "eval_steps_per_second": 0.629,
+      "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,
       "step": 110
     },
     {
       "epoch": 5.454545454545454,
-      "grad_norm": 18.131729125976562,
+      "grad_norm": 5.456847667694092,
       "learning_rate": 2.5252525252525256e-05,
-      "loss": 0.2432,
+      "loss": 0.2711,
       "step": 120
     },
     {
       "epoch": 5.909090909090909,
-      "grad_norm": 54.289772033691406,
+      "grad_norm": 11.77868938446045,
       "learning_rate": 2.272727272727273e-05,
-      "loss": 0.2494,
+      "loss": 0.221,
       "step": 130
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.8842443729903537,
-      "eval_f1": 0.8873642285098018,
-      "eval_loss": 0.20758801698684692,
-      "eval_precision": 0.9071323056151893,
-      "eval_recall": 0.8842443729903537,
-      "eval_runtime": 16.0723,
-      "eval_samples_per_second": 19.35,
-      "eval_steps_per_second": 0.622,
+      "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,
       "step": 132
     },
     {
       "epoch": 6.363636363636363,
-      "grad_norm": 28.0532283782959,
+      "grad_norm": 10.365585327148438,
       "learning_rate": 2.0202020202020203e-05,
-      "loss": 0.2328,
+      "loss": 0.2205,
       "step": 140
     },
     {
       "epoch": 6.818181818181818,
-      "grad_norm": 7.956657886505127,
+      "grad_norm": 3.4123003482818604,
       "learning_rate": 1.7676767676767676e-05,
-      "loss": 0.2239,
+      "loss": 0.2339,
       "step": 150
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.8360128617363344,
-      "eval_f1": 0.8179357935850551,
-      "eval_loss": 0.2842709422111511,
-      "eval_precision": 0.8632228058178503,
-      "eval_recall": 0.8360128617363344,
-      "eval_runtime": 15.7132,
-      "eval_samples_per_second": 19.792,
-      "eval_steps_per_second": 0.636,
+      "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,
       "step": 154
     },
     {
       "epoch": 7.2727272727272725,
-      "grad_norm": 12.15099048614502,
+      "grad_norm": 6.866532325744629,
       "learning_rate": 1.5151515151515153e-05,
-      "loss": 0.2387,
+      "loss": 0.2274,
       "step": 160
     },
     {
       "epoch": 7.7272727272727275,
-      "grad_norm": 3.2118804454803467,
+      "grad_norm": 28.103683471679688,
       "learning_rate": 1.2626262626262628e-05,
-      "loss": 0.2338,
+      "loss": 0.2215,
       "step": 170
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.842443729903537,
-      "eval_f1": 0.8309118837315677,
-      "eval_loss": 0.3346959054470062,
-      "eval_precision": 0.8517411995396645,
-      "eval_recall": 0.842443729903537,
-      "eval_runtime": 15.7304,
-      "eval_samples_per_second": 19.771,
-      "eval_steps_per_second": 0.636,
+      "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,
       "step": 176
     },
     {
       "epoch": 8.181818181818182,
-      "grad_norm": 5.065102577209473,
+      "grad_norm": 7.824560642242432,
       "learning_rate": 1.0101010101010101e-05,
-      "loss": 0.2258,
+      "loss": 0.2202,
       "step": 180
     },
     {
       "epoch": 8.636363636363637,
-      "grad_norm": 5.690997123718262,
+      "grad_norm": 1.545398473739624,
       "learning_rate": 7.5757575757575764e-06,
-      "loss": 0.2163,
+      "loss": 0.2162,
       "step": 190
     },
     {
       "epoch": 9.0,
-      "eval_accuracy": 0.8488745980707395,
-      "eval_f1": 0.8409186699599422,
-      "eval_loss": 0.280775785446167,
-      "eval_precision": 0.8518186827287411,
-      "eval_recall": 0.8488745980707395,
-      "eval_runtime": 15.5946,
-      "eval_samples_per_second": 19.943,
-      "eval_steps_per_second": 0.641,
+      "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,
       "step": 198
     },
     {
       "epoch": 9.090909090909092,
-      "grad_norm": 2.9396848678588867,
+      "grad_norm": 2.2911288738250732,
       "learning_rate": 5.050505050505051e-06,
-      "loss": 0.2183,
+      "loss": 0.2224,
       "step": 200
     },
     {
       "epoch": 9.545454545454545,
-      "grad_norm": 7.1536149978637695,
+      "grad_norm": 3.872638463973999,
       "learning_rate": 2.5252525252525253e-06,
-      "loss": 0.2127,
+      "loss": 0.2126,
       "step": 210
     },
     {
       "epoch": 10.0,
-      "grad_norm": 3.191535711288452,
+      "grad_norm": 2.078859567642212,
       "learning_rate": 0.0,
-      "loss": 0.2303,
+      "loss": 0.2226,
       "step": 220
     },
     {
       "epoch": 10.0,
-      "eval_accuracy": 0.842443729903537,
-      "eval_f1": 0.8355956465755988,
-      "eval_loss": 0.24919378757476807,
-      "eval_precision": 0.8421848602103656,
-      "eval_recall": 0.842443729903537,
-      "eval_runtime": 16.6322,
-      "eval_samples_per_second": 18.699,
-      "eval_steps_per_second": 0.601,
+      "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,
       "step": 220
     },
     {
       "epoch": 10.0,
       "step": 220,
       "total_flos": 6.954705718242509e+17,
-      "train_loss": 0.2417184677991,
-      "train_runtime": 1633.5967,
-      "train_samples_per_second": 17.128,
-      "train_steps_per_second": 0.135
+      "train_loss": 0.23402113264257257,
+      "train_runtime": 1621.2514,
+      "train_samples_per_second": 17.258,
+      "train_steps_per_second": 0.136
     }
   ],
   "logging_steps": 10,