Training in progress, epoch 1

all_results.json

@@ -1,16 +1,8 @@
 {
-    "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,
-    "total_flos": 6.968127952294871e+17,
-    "train_loss": 0.13535271666266702,
-    "train_runtime": 2769.4739,
-    "train_samples_per_second": 10.233,
-    "train_steps_per_second": 0.079
+    "epoch": 10.0,
+    "total_flos": 6.954705718242509e+17,
+    "train_loss": 0.305623214895075,
+    "train_runtime": 2743.5676,
+    "train_samples_per_second": 10.198,
+    "train_steps_per_second": 0.08
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cb9c9fac4b2c97fed5accee50fa055acaf8199e642464afde6e4fa5fce301ee5
+oid sha256:22f51cf4d145343a4023a23ca8eb86f80547edca6ab48652506dc0633f1d4a52
 size 110342832

runs/Oct26_08-08-51_c3ee2852f8cc/events.out.tfevents.1729933029.c3ee2852f8cc.307.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f53f68c522971e02ac40c7bad9ee485137683b69eb53604b031ddd33a5f476cd
+size 6323

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 9.887640449438202,
-    "total_flos": 6.968127952294871e+17,
-    "train_loss": 0.13535271666266702,
-    "train_runtime": 2769.4739,
-    "train_samples_per_second": 10.233,
-    "train_steps_per_second": 0.079
+    "epoch": 10.0,
+    "total_flos": 6.954705718242509e+17,
+    "train_loss": 0.305623214895075,
+    "train_runtime": 2743.5676,
+    "train_samples_per_second": 10.198,
+    "train_steps_per_second": 0.08
 }

trainer_state.json

@@ -1,7 +1,7 @@
 {
-  "best_metric": 0.9695078031212484,
-  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-111",
-  "epoch": 9.887640449438202,
+  "best_metric": 0.8535053727979772,
+  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-110",
+  "epoch": 10.0,
   "eval_steps": 500,
   "global_step": 220,
   "is_hyper_param_search": false,
@@ -9,287 +9,287 @@
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.449438202247191,
-      "grad_norm": 6.729045391082764,
+      "epoch": 0.45454545454545453,
+      "grad_norm": 6.745112895965576,
       "learning_rate": 2.272727272727273e-05,
-      "loss": 0.6016,
+      "loss": 0.6636,
       "step": 10
     },
     {
-      "epoch": 0.898876404494382,
-      "grad_norm": 5.153387546539307,
+      "epoch": 0.9090909090909091,
+      "grad_norm": 20.618379592895508,
       "learning_rate": 4.545454545454546e-05,
-      "loss": 0.2262,
+      "loss": 0.5603,
       "step": 20
     },
     {
-      "epoch": 0.9887640449438202,
-      "eval_accuracy": 0.9365079365079365,
-      "eval_f1": 0.9058027582617746,
-      "eval_loss": 0.20612730085849762,
-      "eval_precision": 0.8770471151423532,
-      "eval_recall": 0.9365079365079365,
-      "eval_runtime": 29.0895,
-      "eval_samples_per_second": 10.829,
-      "eval_steps_per_second": 0.344,
+      "epoch": 1.0,
+      "eval_accuracy": 0.6816720257234726,
+      "eval_f1": 0.5559040132339913,
+      "eval_loss": 0.5609580278396606,
+      "eval_precision": 0.7833316046053315,
+      "eval_recall": 0.6816720257234726,
+      "eval_runtime": 114.3233,
+      "eval_samples_per_second": 2.72,
+      "eval_steps_per_second": 0.087,
       "step": 22
     },
     {
-      "epoch": 1.348314606741573,
-      "grad_norm": 2.6891238689422607,
+      "epoch": 1.3636363636363638,
+      "grad_norm": 26.414987564086914,
       "learning_rate": 4.797979797979798e-05,
-      "loss": 0.1932,
+      "loss": 0.4589,
       "step": 30
     },
     {
-      "epoch": 1.797752808988764,
-      "grad_norm": 10.437292098999023,
+      "epoch": 1.8181818181818183,
+      "grad_norm": 67.25880432128906,
       "learning_rate": 4.545454545454546e-05,
-      "loss": 0.1666,
+      "loss": 0.3934,
       "step": 40
     },
     {
-      "epoch": 1.9775280898876404,
-      "eval_accuracy": 0.9333333333333333,
-      "eval_f1": 0.9042145593869733,
-      "eval_loss": 0.12735532224178314,
-      "eval_precision": 0.8768577494692145,
-      "eval_recall": 0.9333333333333333,
-      "eval_runtime": 28.9476,
-      "eval_samples_per_second": 10.882,
-      "eval_steps_per_second": 0.345,
+      "epoch": 2.0,
+      "eval_accuracy": 0.7909967845659164,
+      "eval_f1": 0.7768331822733873,
+      "eval_loss": 0.39760062098503113,
+      "eval_precision": 0.7878999937662005,
+      "eval_recall": 0.7909967845659164,
+      "eval_runtime": 15.7768,
+      "eval_samples_per_second": 19.713,
+      "eval_steps_per_second": 0.634,
       "step": 44
     },
     {
-      "epoch": 2.247191011235955,
-      "grad_norm": 13.23202896118164,
+      "epoch": 2.2727272727272725,
+      "grad_norm": 27.342533111572266,
       "learning_rate": 4.292929292929293e-05,
-      "loss": 0.1408,
+      "loss": 0.3475,
       "step": 50
     },
     {
-      "epoch": 2.696629213483146,
-      "grad_norm": 13.037389755249023,
+      "epoch": 2.7272727272727275,
+      "grad_norm": 20.105819702148438,
       "learning_rate": 4.0404040404040405e-05,
-      "loss": 0.1168,
+      "loss": 0.2992,
       "step": 60
     },
     {
-      "epoch": 2.966292134831461,
-      "eval_accuracy": 0.9523809523809523,
-      "eval_f1": 0.9437855353283663,
-      "eval_loss": 0.10540811717510223,
-      "eval_precision": 0.9461043676729951,
-      "eval_recall": 0.9523809523809523,
-      "eval_runtime": 28.9757,
-      "eval_samples_per_second": 10.871,
-      "eval_steps_per_second": 0.345,
+      "epoch": 3.0,
+      "eval_accuracy": 0.8070739549839229,
+      "eval_f1": 0.7964534816572959,
+      "eval_loss": 0.30361202359199524,
+      "eval_precision": 0.8044150749918517,
+      "eval_recall": 0.8070739549839229,
+      "eval_runtime": 15.7833,
+      "eval_samples_per_second": 19.704,
+      "eval_steps_per_second": 0.634,
       "step": 66
     },
     {
-      "epoch": 3.146067415730337,
-      "grad_norm": 13.592888832092285,
+      "epoch": 3.1818181818181817,
+      "grad_norm": 18.851722717285156,
       "learning_rate": 3.787878787878788e-05,
-      "loss": 0.132,
+      "loss": 0.299,
       "step": 70
     },
     {
-      "epoch": 3.595505617977528,
-      "grad_norm": 4.916988849639893,
+      "epoch": 3.6363636363636362,
+      "grad_norm": 11.8802490234375,
       "learning_rate": 3.535353535353535e-05,
-      "loss": 0.0984,
+      "loss": 0.2746,
       "step": 80
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.9619047619047619,
-      "eval_f1": 0.9599166265832932,
-      "eval_loss": 0.08242347091436386,
-      "eval_precision": 0.9590699155916548,
-      "eval_recall": 0.9619047619047619,
-      "eval_runtime": 29.2105,
-      "eval_samples_per_second": 10.784,
-      "eval_steps_per_second": 0.342,
-      "step": 89
-    },
-    {
-      "epoch": 4.044943820224719,
-      "grad_norm": 5.384190082550049,
+      "eval_accuracy": 0.7877813504823151,
+      "eval_f1": 0.7799243840827713,
+      "eval_loss": 0.35377901792526245,
+      "eval_precision": 0.7811877113589711,
+      "eval_recall": 0.7877813504823151,
+      "eval_runtime": 15.7385,
+      "eval_samples_per_second": 19.76,
+      "eval_steps_per_second": 0.635,
+      "step": 88
+    },
+    {
+      "epoch": 4.090909090909091,
+      "grad_norm": 28.75580596923828,
       "learning_rate": 3.282828282828283e-05,
-      "loss": 0.0997,
+      "loss": 0.2718,
       "step": 90
     },
     {
-      "epoch": 4.49438202247191,
-      "grad_norm": 6.7096028327941895,
+      "epoch": 4.545454545454545,
+      "grad_norm": 20.20217514038086,
       "learning_rate": 3.0303030303030306e-05,
-      "loss": 0.1319,
+      "loss": 0.275,
       "step": 100
     },
     {
-      "epoch": 4.943820224719101,
-      "grad_norm": 3.9921090602874756,
+      "epoch": 5.0,
+      "grad_norm": 4.023470878601074,
       "learning_rate": 2.777777777777778e-05,
-      "loss": 0.1028,
+      "loss": 0.2573,
       "step": 110
     },
     {
-      "epoch": 4.98876404494382,
-      "eval_accuracy": 0.9714285714285714,
-      "eval_f1": 0.9695078031212484,
-      "eval_loss": 0.08143213391304016,
-      "eval_precision": 0.9696825396825397,
-      "eval_recall": 0.9714285714285714,
-      "eval_runtime": 29.3091,
-      "eval_samples_per_second": 10.748,
-      "eval_steps_per_second": 0.341,
-      "step": 111
+      "epoch": 5.0,
+      "eval_accuracy": 0.8520900321543409,
+      "eval_f1": 0.8535053727979772,
+      "eval_loss": 0.22417089343070984,
+      "eval_precision": 0.8561132846530342,
+      "eval_recall": 0.8520900321543409,
+      "eval_runtime": 16.34,
+      "eval_samples_per_second": 19.033,
+      "eval_steps_per_second": 0.612,
+      "step": 110
     },
     {
-      "epoch": 5.393258426966292,
-      "grad_norm": 7.567589282989502,
+      "epoch": 5.454545454545454,
+      "grad_norm": 10.749618530273438,
       "learning_rate": 2.5252525252525256e-05,
-      "loss": 0.0806,
+      "loss": 0.2656,
       "step": 120
     },
     {
-      "epoch": 5.842696629213483,
-      "grad_norm": 3.001021385192871,
+      "epoch": 5.909090909090909,
+      "grad_norm": 18.935693740844727,
       "learning_rate": 2.272727272727273e-05,
-      "loss": 0.1082,
+      "loss": 0.2724,
       "step": 130
     },
     {
-      "epoch": 5.97752808988764,
-      "eval_accuracy": 0.9492063492063492,
-      "eval_f1": 0.9329456755199329,
-      "eval_loss": 0.08350013196468353,
-      "eval_precision": 0.9518195273822283,
-      "eval_recall": 0.9492063492063492,
-      "eval_runtime": 28.3828,
-      "eval_samples_per_second": 11.098,
-      "eval_steps_per_second": 0.352,
-      "step": 133
+      "epoch": 6.0,
+      "eval_accuracy": 0.77491961414791,
+      "eval_f1": 0.7347099697338514,
+      "eval_loss": 0.38007885217666626,
+      "eval_precision": 0.8145331611697926,
+      "eval_recall": 0.77491961414791,
+      "eval_runtime": 17.438,
+      "eval_samples_per_second": 17.835,
+      "eval_steps_per_second": 0.573,
+      "step": 132
     },
     {
-      "epoch": 6.292134831460674,
-      "grad_norm": 3.2272369861602783,
+      "epoch": 6.363636363636363,
+      "grad_norm": 11.04854679107666,
       "learning_rate": 2.0202020202020203e-05,
-      "loss": 0.0914,
+      "loss": 0.2445,
       "step": 140
     },
     {
-      "epoch": 6.741573033707866,
-      "grad_norm": 11.393094062805176,
+      "epoch": 6.818181818181818,
+      "grad_norm": 5.090071201324463,
       "learning_rate": 1.7676767676767676e-05,
-      "loss": 0.0962,
+      "loss": 0.2344,
       "step": 150
     },
     {
-      "epoch": 6.966292134831461,
-      "eval_accuracy": 0.9587301587301588,
-      "eval_f1": 0.9582359734643997,
-      "eval_loss": 0.08720001578330994,
-      "eval_precision": 0.9578027275395696,
-      "eval_recall": 0.9587301587301588,
-      "eval_runtime": 28.5518,
-      "eval_samples_per_second": 11.033,
-      "eval_steps_per_second": 0.35,
-      "step": 155
+      "epoch": 7.0,
+      "eval_accuracy": 0.8231511254019293,
+      "eval_f1": 0.801136511467,
+      "eval_loss": 0.332674115896225,
+      "eval_precision": 0.8543834389596174,
+      "eval_recall": 0.8231511254019293,
+      "eval_runtime": 15.8127,
+      "eval_samples_per_second": 19.668,
+      "eval_steps_per_second": 0.632,
+      "step": 154
     },
     {
-      "epoch": 7.191011235955056,
-      "grad_norm": 3.3998546600341797,
+      "epoch": 7.2727272727272725,
+      "grad_norm": 6.6000776290893555,
       "learning_rate": 1.5151515151515153e-05,
-      "loss": 0.1013,
+      "loss": 0.241,
       "step": 160
     },
     {
-      "epoch": 7.640449438202247,
-      "grad_norm": 2.875291109085083,
+      "epoch": 7.7272727272727275,
+      "grad_norm": 3.889106035232544,
       "learning_rate": 1.2626262626262628e-05,
-      "loss": 0.0799,
+      "loss": 0.2225,
       "step": 170
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.9587301587301588,
-      "eval_f1": 0.9545950048462611,
-      "eval_loss": 0.08030053973197937,
-      "eval_precision": 0.9543377887748749,
-      "eval_recall": 0.9587301587301588,
-      "eval_runtime": 28.171,
-      "eval_samples_per_second": 11.182,
-      "eval_steps_per_second": 0.355,
-      "step": 178
-    },
-    {
-      "epoch": 8.089887640449438,
-      "grad_norm": 3.642427682876587,
+      "eval_accuracy": 0.8392282958199357,
+      "eval_f1": 0.8220529088336219,
+      "eval_loss": 0.3736245632171631,
+      "eval_precision": 0.8654598204437431,
+      "eval_recall": 0.8392282958199357,
+      "eval_runtime": 15.8357,
+      "eval_samples_per_second": 19.639,
+      "eval_steps_per_second": 0.631,
+      "step": 176
+    },
+    {
+      "epoch": 8.181818181818182,
+      "grad_norm": 11.122193336486816,
       "learning_rate": 1.0101010101010101e-05,
-      "loss": 0.0703,
+      "loss": 0.2323,
       "step": 180
     },
     {
-      "epoch": 8.539325842696629,
-      "grad_norm": 11.242568016052246,
+      "epoch": 8.636363636363637,
+      "grad_norm": 35.049468994140625,
       "learning_rate": 7.5757575757575764e-06,
-      "loss": 0.0869,
+      "loss": 0.225,
       "step": 190
     },
     {
-      "epoch": 8.98876404494382,
-      "grad_norm": 11.542448997497559,
-      "learning_rate": 5.050505050505051e-06,
-      "loss": 0.0954,
-      "step": 200
+      "epoch": 9.0,
+      "eval_accuracy": 0.8327974276527331,
+      "eval_f1": 0.8137864007121186,
+      "eval_loss": 0.3479458689689636,
+      "eval_precision": 0.860997154156041,
+      "eval_recall": 0.8327974276527331,
+      "eval_runtime": 16.0275,
+      "eval_samples_per_second": 19.404,
+      "eval_steps_per_second": 0.624,
+      "step": 198
     },
     {
-      "epoch": 8.98876404494382,
-      "eval_accuracy": 0.9619047619047619,
-      "eval_f1": 0.9587351719211175,
-      "eval_loss": 0.06851905584335327,
-      "eval_precision": 0.9583926593893372,
-      "eval_recall": 0.9619047619047619,
-      "eval_runtime": 28.2782,
-      "eval_samples_per_second": 11.139,
-      "eval_steps_per_second": 0.354,
+      "epoch": 9.090909090909092,
+      "grad_norm": 2.3715155124664307,
+      "learning_rate": 5.050505050505051e-06,
+      "loss": 0.2363,
       "step": 200
     },
     {
-      "epoch": 9.438202247191011,
-      "grad_norm": 4.964309215545654,
+      "epoch": 9.545454545454545,
+      "grad_norm": 4.342611789703369,
       "learning_rate": 2.5252525252525253e-06,
-      "loss": 0.0806,
+      "loss": 0.2184,
       "step": 210
     },
     {
-      "epoch": 9.887640449438202,
-      "grad_norm": 3.8892452716827393,
+      "epoch": 10.0,
+      "grad_norm": 11.119804382324219,
       "learning_rate": 0.0,
-      "loss": 0.0771,
+      "loss": 0.2308,
       "step": 220
     },
     {
-      "epoch": 9.887640449438202,
-      "eval_accuracy": 0.9619047619047619,
-      "eval_f1": 0.9587351719211175,
-      "eval_loss": 0.07111740857362747,
-      "eval_precision": 0.9583926593893372,
-      "eval_recall": 0.9619047619047619,
-      "eval_runtime": 28.6883,
-      "eval_samples_per_second": 10.98,
-      "eval_steps_per_second": 0.349,
+      "epoch": 10.0,
+      "eval_accuracy": 0.8327974276527331,
+      "eval_f1": 0.8137864007121186,
+      "eval_loss": 0.32772162556648254,
+      "eval_precision": 0.860997154156041,
+      "eval_recall": 0.8327974276527331,
+      "eval_runtime": 17.0105,
+      "eval_samples_per_second": 18.283,
+      "eval_steps_per_second": 0.588,
       "step": 220
     },
     {
-      "epoch": 9.887640449438202,
+      "epoch": 10.0,
       "step": 220,
-      "total_flos": 6.968127952294871e+17,
-      "train_loss": 0.13535271666266702,
-      "train_runtime": 2769.4739,
-      "train_samples_per_second": 10.233,
-      "train_steps_per_second": 0.079
+      "total_flos": 6.954705718242509e+17,
+      "train_loss": 0.305623214895075,
+      "train_runtime": 2743.5676,
+      "train_samples_per_second": 10.198,
+      "train_steps_per_second": 0.08
     }
   ],
   "logging_steps": 10,
@@ -309,7 +309,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 6.968127952294871e+17,
+  "total_flos": 6.954705718242509e+17,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": null