End of training

all_results.json

@@ -1,12 +1,12 @@
 {
     "epoch": 3.0,
-    "eval_accuracy": 0.9825148809523809,
-    "eval_loss": 0.1412353515625,
-    "eval_runtime": 8.1951,
-    "eval_samples_per_second": 328.002,
-    "eval_steps_per_second": 20.5,
-    "train_loss": 0.46718859798693785,
-    "train_runtime": 815.2101,
-    "train_samples_per_second": 89.027,
-    "train_steps_per_second": 0.696
+    "eval_accuracy": 0.9791666666666666,
+    "eval_loss": 0.1474609375,
+    "eval_runtime": 7.6628,
+    "eval_samples_per_second": 350.786,
+    "eval_steps_per_second": 21.924,
+    "train_loss": 0.47797061442487876,
+    "train_runtime": 780.2883,
+    "train_samples_per_second": 93.012,
+    "train_steps_per_second": 0.727
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 3.0,
-    "eval_accuracy": 0.9825148809523809,
-    "eval_loss": 0.1412353515625,
-    "eval_runtime": 8.1951,
-    "eval_samples_per_second": 328.002,
-    "eval_steps_per_second": 20.5
+    "eval_accuracy": 0.9791666666666666,
+    "eval_loss": 0.1474609375,
+    "eval_runtime": 7.6628,
+    "eval_samples_per_second": 350.786,
+    "eval_steps_per_second": 21.924
 }

train_results.json

@@ -1,7 +1,7 @@
 {
     "epoch": 3.0,
-    "train_loss": 0.46718859798693785,
-    "train_runtime": 815.2101,
-    "train_samples_per_second": 89.027,
-    "train_steps_per_second": 0.696
+    "train_loss": 0.47797061442487876,
+    "train_runtime": 780.2883,
+    "train_samples_per_second": 93.012,
+    "train_steps_per_second": 0.727
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.9825148809523809,
+  "best_metric": 0.9791666666666666,
   "best_model_checkpoint": "vit-base-patch16-224-in21k-finetuned-eurosat/checkpoint-567",
   "epoch": 3.0,
   "global_step": 567,
@@ -10,379 +10,379 @@
     {
       "epoch": 0.05,
       "learning_rate": 8.771929824561403e-06,
-      "loss": 2.3141,
+      "loss": 2.3127,
       "step": 10
     },
     {
       "epoch": 0.11,
       "learning_rate": 1.7543859649122806e-05,
-      "loss": 2.2088,
+      "loss": 2.2527,
       "step": 20
     },
     {
       "epoch": 0.16,
       "learning_rate": 2.6315789473684212e-05,
-      "loss": 2.1541,
+      "loss": 2.1359,
       "step": 30
     },
     {
       "epoch": 0.21,
       "learning_rate": 3.508771929824561e-05,
-      "loss": 1.8185,
+      "loss": 1.8139,
       "step": 40
     },
     {
       "epoch": 0.26,
       "learning_rate": 4.3859649122807014e-05,
-      "loss": 1.468,
+      "loss": 1.5964,
       "step": 50
     },
     {
       "epoch": 0.32,
       "learning_rate": 4.970588235294118e-05,
-      "loss": 1.3207,
+      "loss": 1.2502,
       "step": 60
     },
     {
       "epoch": 0.37,
       "learning_rate": 4.872549019607843e-05,
-      "loss": 0.9924,
+      "loss": 0.962,
       "step": 70
     },
     {
       "epoch": 0.42,
       "learning_rate": 4.774509803921569e-05,
-      "loss": 0.7575,
+      "loss": 0.8233,
       "step": 80
     },
     {
       "epoch": 0.48,
       "learning_rate": 4.6764705882352944e-05,
-      "loss": 0.4987,
+      "loss": 0.5574,
       "step": 90
     },
     {
       "epoch": 0.53,
       "learning_rate": 4.5784313725490194e-05,
-      "loss": 0.5602,
+      "loss": 0.6041,
       "step": 100
     },
     {
       "epoch": 0.58,
       "learning_rate": 4.480392156862745e-05,
-      "loss": 0.6816,
+      "loss": 0.484,
       "step": 110
     },
     {
       "epoch": 0.63,
       "learning_rate": 4.382352941176471e-05,
-      "loss": 0.6875,
+      "loss": 0.4123,
       "step": 120
     },
     {
       "epoch": 0.69,
       "learning_rate": 4.284313725490196e-05,
-      "loss": 0.4998,
+      "loss": 0.3052,
       "step": 130
     },
     {
       "epoch": 0.74,
       "learning_rate": 4.1862745098039217e-05,
-      "loss": 0.4879,
+      "loss": 0.5792,
       "step": 140
     },
     {
       "epoch": 0.79,
       "learning_rate": 4.0882352941176474e-05,
-      "loss": 0.3393,
+      "loss": 0.6366,
       "step": 150
     },
     {
       "epoch": 0.85,
       "learning_rate": 3.990196078431373e-05,
-      "loss": 0.2399,
+      "loss": 0.3072,
       "step": 160
     },
     {
       "epoch": 0.9,
       "learning_rate": 3.892156862745098e-05,
-      "loss": 0.3542,
+      "loss": 0.506,
       "step": 170
     },
     {
       "epoch": 0.95,
       "learning_rate": 3.794117647058824e-05,
-      "loss": 0.2284,
+      "loss": 0.3961,
       "step": 180
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.9657738095238095,
-      "eval_loss": 0.27978515625,
-      "eval_runtime": 7.7904,
-      "eval_samples_per_second": 345.039,
-      "eval_steps_per_second": 21.565,
+      "eval_accuracy": 0.9609375,
+      "eval_loss": 0.29638671875,
+      "eval_runtime": 7.8374,
+      "eval_samples_per_second": 342.973,
+      "eval_steps_per_second": 21.436,
       "step": 189
     },
     {
       "epoch": 1.01,
       "learning_rate": 3.6960784313725496e-05,
-      "loss": 0.2552,
+      "loss": 0.3134,
       "step": 190
     },
     {
       "epoch": 1.06,
       "learning_rate": 3.5980392156862746e-05,
-      "loss": 0.2004,
+      "loss": 0.301,
       "step": 200
     },
     {
       "epoch": 1.11,
       "learning_rate": 3.5e-05,
-      "loss": 0.2532,
+      "loss": 0.2332,
       "step": 210
     },
     {
       "epoch": 1.16,
       "learning_rate": 3.401960784313726e-05,
-      "loss": 0.3075,
+      "loss": 0.2439,
       "step": 220
     },
     {
       "epoch": 1.22,
       "learning_rate": 3.303921568627451e-05,
-      "loss": 0.2281,
+      "loss": 0.3456,
       "step": 230
     },
     {
       "epoch": 1.27,
       "learning_rate": 3.205882352941177e-05,
-      "loss": 0.3189,
+      "loss": 0.4117,
       "step": 240
     },
     {
       "epoch": 1.32,
       "learning_rate": 3.107843137254902e-05,
-      "loss": 0.3038,
+      "loss": 0.2915,
       "step": 250
     },
     {
       "epoch": 1.38,
       "learning_rate": 3.0098039215686273e-05,
-      "loss": 0.1736,
+      "loss": 0.3343,
       "step": 260
     },
     {
       "epoch": 1.43,
       "learning_rate": 2.9117647058823534e-05,
-      "loss": 0.2718,
+      "loss": 0.4646,
       "step": 270
     },
     {
       "epoch": 1.48,
       "learning_rate": 2.8137254901960784e-05,
-      "loss": 0.3009,
+      "loss": 0.2341,
       "step": 280
     },
     {
       "epoch": 1.53,
       "learning_rate": 2.7156862745098038e-05,
-      "loss": 0.2359,
+      "loss": 0.2378,
       "step": 290
     },
     {
       "epoch": 1.59,
       "learning_rate": 2.6176470588235295e-05,
-      "loss": 0.2563,
+      "loss": 0.1518,
       "step": 300
     },
     {
       "epoch": 1.64,
       "learning_rate": 2.519607843137255e-05,
-      "loss": 0.2496,
+      "loss": 0.1586,
       "step": 310
     },
     {
       "epoch": 1.69,
       "learning_rate": 2.4215686274509806e-05,
-      "loss": 0.324,
+      "loss": 0.2091,
       "step": 320
     },
     {
       "epoch": 1.75,
       "learning_rate": 2.323529411764706e-05,
-      "loss": 0.2806,
+      "loss": 0.3397,
       "step": 330
     },
     {
       "epoch": 1.8,
       "learning_rate": 2.2254901960784314e-05,
-      "loss": 0.3569,
+      "loss": 0.3218,
       "step": 340
     },
     {
       "epoch": 1.85,
       "learning_rate": 2.1274509803921568e-05,
-      "loss": 0.2734,
+      "loss": 0.1686,
       "step": 350
     },
     {
       "epoch": 1.9,
       "learning_rate": 2.0294117647058825e-05,
-      "loss": 0.2582,
+      "loss": 0.2736,
       "step": 360
     },
     {
       "epoch": 1.96,
       "learning_rate": 1.931372549019608e-05,
-      "loss": 0.2262,
+      "loss": 0.3819,
       "step": 370
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.9799107142857143,
-      "eval_loss": 0.1624755859375,
-      "eval_runtime": 7.8583,
-      "eval_samples_per_second": 342.057,
-      "eval_steps_per_second": 21.379,
+      "eval_accuracy": 0.9773065476190477,
+      "eval_loss": 0.17041015625,
+      "eval_runtime": 7.4088,
+      "eval_samples_per_second": 362.809,
+      "eval_steps_per_second": 22.676,
       "step": 378
     },
     {
       "epoch": 2.01,
       "learning_rate": 1.8333333333333333e-05,
-      "loss": 0.4627,
+      "loss": 0.1427,
       "step": 380
     },
     {
       "epoch": 2.06,
       "learning_rate": 1.735294117647059e-05,
-      "loss": 0.2553,
+      "loss": 0.3301,
       "step": 390
     },
     {
       "epoch": 2.12,
       "learning_rate": 1.6372549019607844e-05,
-      "loss": 0.1291,
+      "loss": 0.144,
       "step": 400
     },
     {
       "epoch": 2.17,
       "learning_rate": 1.53921568627451e-05,
-      "loss": 0.1432,
+      "loss": 0.2526,
       "step": 410
     },
     {
       "epoch": 2.22,
       "learning_rate": 1.4411764705882352e-05,
-      "loss": 0.3603,
+      "loss": 0.2282,
       "step": 420
     },
     {
       "epoch": 2.28,
       "learning_rate": 1.3431372549019607e-05,
-      "loss": 0.2984,
+      "loss": 0.2293,
       "step": 430
     },
     {
       "epoch": 2.33,
       "learning_rate": 1.2450980392156863e-05,
-      "loss": 0.1232,
+      "loss": 0.1889,
       "step": 440
     },
     {
       "epoch": 2.38,
       "learning_rate": 1.1470588235294118e-05,
-      "loss": 0.215,
+      "loss": 0.1527,
       "step": 450
     },
     {
       "epoch": 2.43,
       "learning_rate": 1.0490196078431374e-05,
-      "loss": 0.1089,
+      "loss": 0.1131,
       "step": 460
     },
     {
       "epoch": 2.49,
       "learning_rate": 9.509803921568628e-06,
-      "loss": 0.2441,
+      "loss": 0.1541,
       "step": 470
     },
     {
       "epoch": 2.54,
       "learning_rate": 8.529411764705883e-06,
-      "loss": 0.127,
+      "loss": 0.1328,
       "step": 480
     },
     {
       "epoch": 2.59,
       "learning_rate": 7.549019607843137e-06,
-      "loss": 0.1098,
+      "loss": 0.1335,
       "step": 490
     },
     {
       "epoch": 2.65,
       "learning_rate": 6.568627450980393e-06,
-      "loss": 0.1042,
+      "loss": 0.2389,
       "step": 500
     },
     {
       "epoch": 2.7,
       "learning_rate": 5.588235294117647e-06,
-      "loss": 0.1162,
+      "loss": 0.1946,
       "step": 510
     },
     {
       "epoch": 2.75,
       "learning_rate": 4.607843137254902e-06,
-      "loss": 0.1821,
+      "loss": 0.1927,
       "step": 520
     },
     {
       "epoch": 2.8,
       "learning_rate": 3.6274509803921573e-06,
-      "loss": 0.3014,
+      "loss": 0.2805,
       "step": 530
     },
     {
       "epoch": 2.86,
       "learning_rate": 2.647058823529412e-06,
-      "loss": 0.181,
+      "loss": 0.1217,
       "step": 540
     },
     {
       "epoch": 2.91,
       "learning_rate": 1.6666666666666667e-06,
-      "loss": 0.1033,
+      "loss": 0.1965,
       "step": 550
     },
     {
       "epoch": 2.96,
       "learning_rate": 6.862745098039216e-07,
-      "loss": 0.1693,
+      "loss": 0.152,
       "step": 560
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.9825148809523809,
-      "eval_loss": 0.1412353515625,
-      "eval_runtime": 7.3548,
-      "eval_samples_per_second": 365.475,
-      "eval_steps_per_second": 22.842,
+      "eval_accuracy": 0.9791666666666666,
+      "eval_loss": 0.1474609375,
+      "eval_runtime": 7.3981,
+      "eval_samples_per_second": 363.338,
+      "eval_steps_per_second": 22.709,
       "step": 567
     },
     {
       "epoch": 3.0,
       "step": 567,
       "total_flos": 7.199311221168632e+20,
-      "train_loss": 0.46718859798693785,
-      "train_runtime": 815.2101,
-      "train_samples_per_second": 89.027,
-      "train_steps_per_second": 0.696
+      "train_loss": 0.47797061442487876,
+      "train_runtime": 780.2883,
+      "train_samples_per_second": 93.012,
+      "train_steps_per_second": 0.727
     }
   ],
   "max_steps": 567,
   "num_train_epochs": 3,
-  "start_time": 1670273012.1493745,
+  "start_time": 1670358735.7488465,
   "total_flos": 7.199311221168632e+20,
   "trial_name": null,
   "trial_params": null