Training in progress, epoch 1

all_results.json

@@ -6,8 +6,8 @@
     "eval_samples_per_second": 79.633,
     "eval_steps_per_second": 4.977,
     "total_flos": 4.860822854913491e+18,
-    "train_loss": 1.0101308201298569,
-    "train_runtime": 2441.1089,
-    "train_samples_per_second": 8.647,
-    "train_steps_per_second": 0.135
+    "train_loss": 0.3027138883417303,
+    "train_runtime": 956.6189,
+    "train_samples_per_second": 22.065,
+    "train_steps_per_second": 0.345
 }

runs/Jul05_15-01-11_c8d5d7484861/events.out.tfevents.1688574849.c8d5d7484861.2170.5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c8f05691efcb2629cddd73085bc3e001f1dda740e0a3d8d17ee5b11298c8921
+size 5772

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 6.0,
     "total_flos": 4.860822854913491e+18,
-    "train_loss": 1.0101308201298569,
-    "train_runtime": 2441.1089,
-    "train_samples_per_second": 8.647,
-    "train_steps_per_second": 0.135
+    "train_loss": 0.3027138883417303,
+    "train_runtime": 956.6189,
+    "train_samples_per_second": 22.065,
+    "train_steps_per_second": 0.345
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.8795454545454545,
+  "best_metric": 0.9659090909090909,
   "best_model_checkpoint": "swinv2-large-patch4-window12to16-192to256-22kto1k-ft-finetuned-eurosat-50/checkpoint-330",
   "epoch": 6.0,
   "global_step": 330,
@@ -10,263 +10,263 @@
     {
       "epoch": 0.18,
       "learning_rate": 6.060606060606061e-07,
-      "loss": 2.0126,
+      "loss": 0.3989,
       "step": 10
     },
     {
       "epoch": 0.36,
       "learning_rate": 1.2121212121212122e-06,
-      "loss": 2.0149,
+      "loss": 0.4458,
       "step": 20
     },
     {
       "epoch": 0.55,
       "learning_rate": 1.8181818181818183e-06,
-      "loss": 1.9557,
+      "loss": 0.4395,
       "step": 30
     },
     {
       "epoch": 0.73,
       "learning_rate": 2.4242424242424244e-06,
-      "loss": 1.9057,
+      "loss": 0.4139,
       "step": 40
     },
     {
       "epoch": 0.91,
       "learning_rate": 3.0303030303030305e-06,
-      "loss": 1.8498,
+      "loss": 0.4562,
       "step": 50
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.32727272727272727,
-      "eval_loss": 1.734784483909607,
-      "eval_runtime": 359.3738,
-      "eval_samples_per_second": 2.449,
-      "eval_steps_per_second": 0.153,
+      "eval_accuracy": 0.8897727272727273,
+      "eval_loss": 0.3206745386123657,
+      "eval_runtime": 10.9207,
+      "eval_samples_per_second": 80.581,
+      "eval_steps_per_second": 5.036,
       "step": 55
     },
     {
       "epoch": 1.09,
       "learning_rate": 3.6363636363636366e-06,
-      "loss": 1.7745,
+      "loss": 0.424,
       "step": 60
     },
     {
       "epoch": 1.27,
       "learning_rate": 4.242424242424243e-06,
-      "loss": 1.7178,
+      "loss": 0.4218,
       "step": 70
     },
     {
       "epoch": 1.45,
       "learning_rate": 4.848484848484849e-06,
-      "loss": 1.5662,
+      "loss": 0.3912,
       "step": 80
     },
     {
       "epoch": 1.64,
       "learning_rate": 5.4545454545454545e-06,
-      "loss": 1.4263,
+      "loss": 0.3665,
       "step": 90
     },
     {
       "epoch": 1.82,
       "learning_rate": 6.060606060606061e-06,
-      "loss": 1.2797,
+      "loss": 0.3969,
       "step": 100
     },
     {
       "epoch": 2.0,
       "learning_rate": 6.666666666666667e-06,
-      "loss": 1.1886,
+      "loss": 0.372,
       "step": 110
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.6102272727272727,
-      "eval_loss": 1.019785761833191,
-      "eval_runtime": 11.0093,
-      "eval_samples_per_second": 79.933,
-      "eval_steps_per_second": 4.996,
+      "eval_accuracy": 0.9068181818181819,
+      "eval_loss": 0.2666811943054199,
+      "eval_runtime": 10.9113,
+      "eval_samples_per_second": 80.65,
+      "eval_steps_per_second": 5.041,
       "step": 110
     },
     {
       "epoch": 2.18,
       "learning_rate": 7.272727272727273e-06,
-      "loss": 1.099,
+      "loss": 0.35,
       "step": 120
     },
     {
       "epoch": 2.36,
       "learning_rate": 7.87878787878788e-06,
-      "loss": 1.0145,
+      "loss": 0.275,
       "step": 130
     },
     {
       "epoch": 2.55,
       "learning_rate": 8.484848484848486e-06,
-      "loss": 0.9682,
+      "loss": 0.2911,
       "step": 140
     },
     {
       "epoch": 2.73,
       "learning_rate": 9.090909090909091e-06,
-      "loss": 0.9164,
+      "loss": 0.2794,
       "step": 150
     },
     {
       "epoch": 2.91,
       "learning_rate": 9.696969696969698e-06,
-      "loss": 0.8636,
+      "loss": 0.2776,
       "step": 160
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.7397727272727272,
-      "eval_loss": 0.6859003305435181,
-      "eval_runtime": 10.9928,
-      "eval_samples_per_second": 80.052,
-      "eval_steps_per_second": 5.003,
+      "eval_accuracy": 0.9125,
+      "eval_loss": 0.2862165570259094,
+      "eval_runtime": 11.069,
+      "eval_samples_per_second": 79.501,
+      "eval_steps_per_second": 4.969,
       "step": 165
     },
     {
       "epoch": 3.09,
       "learning_rate": 9.696969696969698e-06,
-      "loss": 0.8709,
+      "loss": 0.3096,
       "step": 170
     },
     {
       "epoch": 3.27,
       "learning_rate": 9.090909090909091e-06,
-      "loss": 0.749,
+      "loss": 0.2071,
       "step": 180
     },
     {
       "epoch": 3.45,
       "learning_rate": 8.484848484848486e-06,
-      "loss": 0.7015,
+      "loss": 0.2234,
       "step": 190
     },
     {
       "epoch": 3.64,
       "learning_rate": 7.87878787878788e-06,
-      "loss": 0.6916,
+      "loss": 0.2446,
       "step": 200
     },
     {
       "epoch": 3.82,
       "learning_rate": 7.272727272727273e-06,
-      "loss": 0.6508,
+      "loss": 0.1926,
       "step": 210
     },
     {
       "epoch": 4.0,
       "learning_rate": 6.666666666666667e-06,
-      "loss": 0.576,
+      "loss": 0.2018,
       "step": 220
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.8477272727272728,
-      "eval_loss": 0.4356772005558014,
-      "eval_runtime": 10.9979,
-      "eval_samples_per_second": 80.015,
-      "eval_steps_per_second": 5.001,
+      "eval_accuracy": 0.9397727272727273,
+      "eval_loss": 0.19657106697559357,
+      "eval_runtime": 10.9849,
+      "eval_samples_per_second": 80.11,
+      "eval_steps_per_second": 5.007,
       "step": 220
     },
     {
       "epoch": 4.18,
       "learning_rate": 6.060606060606061e-06,
-      "loss": 0.51,
+      "loss": 0.1894,
       "step": 230
     },
     {
       "epoch": 4.36,
       "learning_rate": 5.4545454545454545e-06,
-      "loss": 0.5592,
+      "loss": 0.2126,
       "step": 240
     },
     {
       "epoch": 4.55,
       "learning_rate": 4.848484848484849e-06,
-      "loss": 0.5314,
+      "loss": 0.2101,
       "step": 250
     },
     {
       "epoch": 4.73,
       "learning_rate": 4.242424242424243e-06,
-      "loss": 0.5021,
+      "loss": 0.2062,
       "step": 260
     },
     {
       "epoch": 4.91,
       "learning_rate": 3.6363636363636366e-06,
-      "loss": 0.5875,
+      "loss": 0.2751,
       "step": 270
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.8386363636363636,
-      "eval_loss": 0.418759286403656,
-      "eval_runtime": 10.9077,
-      "eval_samples_per_second": 80.677,
-      "eval_steps_per_second": 5.042,
+      "eval_accuracy": 0.9375,
+      "eval_loss": 0.1937328726053238,
+      "eval_runtime": 10.9353,
+      "eval_samples_per_second": 80.474,
+      "eval_steps_per_second": 5.03,
       "step": 275
     },
     {
       "epoch": 5.09,
       "learning_rate": 3.0303030303030305e-06,
-      "loss": 0.4782,
+      "loss": 0.219,
       "step": 280
     },
     {
       "epoch": 5.27,
       "learning_rate": 2.4242424242424244e-06,
-      "loss": 0.521,
+      "loss": 0.2875,
       "step": 290
     },
     {
       "epoch": 5.45,
       "learning_rate": 1.8181818181818183e-06,
-      "loss": 0.451,
+      "loss": 0.2401,
       "step": 300
     },
     {
       "epoch": 5.64,
       "learning_rate": 1.2121212121212122e-06,
-      "loss": 0.4403,
+      "loss": 0.2208,
       "step": 310
     },
     {
       "epoch": 5.82,
       "learning_rate": 6.060606060606061e-07,
-      "loss": 0.4929,
+      "loss": 0.2734,
       "step": 320
     },
     {
       "epoch": 6.0,
       "learning_rate": 0.0,
-      "loss": 0.4677,
+      "loss": 0.2764,
       "step": 330
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.8795454545454545,
-      "eval_loss": 0.3292839825153351,
-      "eval_runtime": 10.9849,
-      "eval_samples_per_second": 80.11,
-      "eval_steps_per_second": 5.007,
+      "eval_accuracy": 0.9659090909090909,
+      "eval_loss": 0.11989254504442215,
+      "eval_runtime": 11.0047,
+      "eval_samples_per_second": 79.966,
+      "eval_steps_per_second": 4.998,
       "step": 330
     },
     {
       "epoch": 6.0,
       "step": 330,
       "total_flos": 4.860822854913491e+18,
-      "train_loss": 1.0101308201298569,
-      "train_runtime": 2441.1089,
-      "train_samples_per_second": 8.647,
-      "train_steps_per_second": 0.135
+      "train_loss": 0.3027138883417303,
+      "train_runtime": 956.6189,
+      "train_samples_per_second": 22.065,
+      "train_steps_per_second": 0.345
     }
   ],
   "max_steps": 330,