End of training

all_results.json

@@ -1,16 +1,16 @@
 {
-    "epoch": 4.0,
-    "eval_accuracy": 0.9977843426883308,
-    "eval_f1": 0.9984067976633033,
-    "eval_loss": 0.010051622986793518,
-    "eval_precision": 0.9989373007438895,
-    "eval_recall": 0.9978768577494692,
-    "eval_runtime": 66.6969,
-    "eval_samples_per_second": 20.301,
-    "eval_steps_per_second": 0.645,
-    "total_flos": 3.037401854071603e+17,
-    "train_loss": 0.2525144927203655,
-    "train_runtime": 9271.044,
-    "train_samples_per_second": 1.318,
-    "train_steps_per_second": 0.01
+    "epoch": 3.99,
+    "eval_accuracy": 0.975609756097561,
+    "eval_f1": 0.975609756097561,
+    "eval_loss": 0.08477138727903366,
+    "eval_precision": 0.975609756097561,
+    "eval_recall": 0.975609756097561,
+    "eval_runtime": 59.1943,
+    "eval_samples_per_second": 20.086,
+    "eval_steps_per_second": 0.642,
+    "total_flos": 1.061932048026071e+18,
+    "train_loss": 0.27396771652870866,
+    "train_runtime": 7327.0516,
+    "train_samples_per_second": 5.841,
+    "train_steps_per_second": 0.045
 }

eval_results.json

@@ -1,11 +1,11 @@
 {
-    "epoch": 4.0,
-    "eval_accuracy": 0.9977843426883308,
-    "eval_f1": 0.9984067976633033,
-    "eval_loss": 0.010051622986793518,
-    "eval_precision": 0.9989373007438895,
-    "eval_recall": 0.9978768577494692,
-    "eval_runtime": 66.6969,
-    "eval_samples_per_second": 20.301,
-    "eval_steps_per_second": 0.645
+    "epoch": 3.99,
+    "eval_accuracy": 0.975609756097561,
+    "eval_f1": 0.975609756097561,
+    "eval_loss": 0.08477138727903366,
+    "eval_precision": 0.975609756097561,
+    "eval_recall": 0.975609756097561,
+    "eval_runtime": 59.1943,
+    "eval_samples_per_second": 20.086,
+    "eval_steps_per_second": 0.642
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 4.0,
-    "total_flos": 3.037401854071603e+17,
-    "train_loss": 0.2525144927203655,
-    "train_runtime": 9271.044,
-    "train_samples_per_second": 1.318,
-    "train_steps_per_second": 0.01
+    "epoch": 3.99,
+    "total_flos": 1.061932048026071e+18,
+    "train_loss": 0.27396771652870866,
+    "train_runtime": 7327.0516,
+    "train_samples_per_second": 5.841,
+    "train_steps_per_second": 0.045
 }

trainer_state.json

@@ -1,127 +1,271 @@
 {
-  "best_metric": 0.9911764705882353,
-  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-brainTumorData/checkpoint-96",
-  "epoch": 4.0,
-  "global_step": 96,
+  "best_metric": 0.975609756097561,
+  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-brainTumorData/checkpoint-332",
+  "epoch": 3.991044776119403,
+  "global_step": 332,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.42,
-      "learning_rate": 5e-05,
-      "loss": 0.5984,
+      "epoch": 0.12,
+      "learning_rate": 1.4705882352941177e-05,
+      "loss": 0.3929,
       "step": 10
     },
     {
-      "epoch": 0.83,
-      "learning_rate": 4.418604651162791e-05,
-      "loss": 0.3405,
+      "epoch": 0.24,
+      "learning_rate": 2.9411764705882354e-05,
+      "loss": 0.3941,
       "step": 20
     },
     {
-      "epoch": 1.0,
-      "eval_accuracy": 0.9647058823529412,
-      "eval_f1": 0.96875,
-      "eval_loss": 0.1137252226471901,
-      "eval_precision": 0.9841269841269841,
-      "eval_recall": 0.9538461538461539,
-      "eval_runtime": 147.9637,
-      "eval_samples_per_second": 2.298,
-      "eval_steps_per_second": 0.074,
-      "step": 24
-    },
-    {
-      "epoch": 1.25,
-      "learning_rate": 3.837209302325582e-05,
-      "loss": 0.2572,
+      "epoch": 0.36,
+      "learning_rate": 4.411764705882353e-05,
+      "loss": 0.3932,
       "step": 30
     },
     {
-      "epoch": 1.67,
-      "learning_rate": 3.2558139534883724e-05,
-      "loss": 0.2449,
+      "epoch": 0.48,
+      "learning_rate": 4.8993288590604034e-05,
+      "loss": 0.3336,
       "step": 40
     },
     {
-      "epoch": 2.0,
-      "eval_accuracy": 0.9735294117647059,
-      "eval_f1": 0.9766233766233765,
-      "eval_loss": 0.08108817040920258,
-      "eval_precision": 0.9894736842105263,
-      "eval_recall": 0.9641025641025641,
-      "eval_runtime": 86.8569,
-      "eval_samples_per_second": 3.914,
-      "eval_steps_per_second": 0.127,
-      "step": 48
-    },
-    {
-      "epoch": 2.08,
-      "learning_rate": 2.674418604651163e-05,
-      "loss": 0.2267,
+      "epoch": 0.6,
+      "learning_rate": 4.731543624161074e-05,
+      "loss": 0.2992,
       "step": 50
     },
     {
-      "epoch": 2.5,
-      "learning_rate": 2.0930232558139536e-05,
-      "loss": 0.1601,
+      "epoch": 0.72,
+      "learning_rate": 4.5637583892617453e-05,
+      "loss": 0.2637,
       "step": 60
     },
     {
-      "epoch": 2.92,
-      "learning_rate": 1.5116279069767441e-05,
-      "loss": 0.1713,
+      "epoch": 0.84,
+      "learning_rate": 4.395973154362416e-05,
+      "loss": 0.2686,
       "step": 70
     },
     {
-      "epoch": 3.0,
-      "eval_accuracy": 0.9794117647058823,
-      "eval_f1": 0.9818181818181819,
-      "eval_loss": 0.061323098838329315,
-      "eval_precision": 0.9947368421052631,
-      "eval_recall": 0.9692307692307692,
-      "eval_runtime": 86.512,
-      "eval_samples_per_second": 3.93,
-      "eval_steps_per_second": 0.127,
-      "step": 72
+      "epoch": 0.96,
+      "learning_rate": 4.228187919463087e-05,
+      "loss": 0.3194,
+      "step": 80
     },
     {
-      "epoch": 3.33,
-      "learning_rate": 9.302325581395349e-06,
-      "loss": 0.1536,
-      "step": 80
+      "epoch": 0.99,
+      "eval_accuracy": 0.911690496215307,
+      "eval_f1": 0.911690496215307,
+      "eval_loss": 0.2373594045639038,
+      "eval_precision": 0.911690496215307,
+      "eval_recall": 0.911690496215307,
+      "eval_runtime": 67.1335,
+      "eval_samples_per_second": 17.711,
+      "eval_steps_per_second": 0.566,
+      "step": 83
     },
     {
-      "epoch": 3.75,
-      "learning_rate": 3.488372093023256e-06,
-      "loss": 0.1615,
+      "epoch": 1.08,
+      "learning_rate": 4.060402684563759e-05,
+      "loss": 0.383,
       "step": 90
     },
     {
-      "epoch": 4.0,
-      "eval_accuracy": 0.9911764705882353,
-      "eval_f1": 0.9923273657289001,
-      "eval_loss": 0.04839443415403366,
-      "eval_precision": 0.9897959183673469,
-      "eval_recall": 0.9948717948717949,
-      "eval_runtime": 86.7553,
-      "eval_samples_per_second": 3.919,
-      "eval_steps_per_second": 0.127,
-      "step": 96
-    },
-    {
-      "epoch": 4.0,
-      "step": 96,
-      "total_flos": 3.037401854071603e+17,
-      "train_loss": 0.2525144927203655,
-      "train_runtime": 9271.044,
-      "train_samples_per_second": 1.318,
-      "train_steps_per_second": 0.01
+      "epoch": 1.2,
+      "learning_rate": 3.89261744966443e-05,
+      "loss": 0.3271,
+      "step": 100
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 3.724832214765101e-05,
+      "loss": 0.3048,
+      "step": 110
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 3.557046979865772e-05,
+      "loss": 0.3089,
+      "step": 120
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 3.389261744966443e-05,
+      "loss": 0.2859,
+      "step": 130
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 3.221476510067114e-05,
+      "loss": 0.2885,
+      "step": 140
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 3.0536912751677856e-05,
+      "loss": 0.2691,
+      "step": 150
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 2.885906040268457e-05,
+      "loss": 0.2871,
+      "step": 160
+    },
+    {
+      "epoch": 1.99,
+      "eval_accuracy": 0.9596299411269975,
+      "eval_f1": 0.9596299411269975,
+      "eval_loss": 0.11854199320077896,
+      "eval_precision": 0.9596299411269975,
+      "eval_recall": 0.9596299411269975,
+      "eval_runtime": 80.2509,
+      "eval_samples_per_second": 14.816,
+      "eval_steps_per_second": 0.474,
+      "step": 166
+    },
+    {
+      "epoch": 2.05,
+      "learning_rate": 2.7181208053691276e-05,
+      "loss": 0.276,
+      "step": 170
+    },
+    {
+      "epoch": 2.17,
+      "learning_rate": 2.550335570469799e-05,
+      "loss": 0.2644,
+      "step": 180
+    },
+    {
+      "epoch": 2.29,
+      "learning_rate": 2.38255033557047e-05,
+      "loss": 0.2082,
+      "step": 190
+    },
+    {
+      "epoch": 2.41,
+      "learning_rate": 2.2147651006711412e-05,
+      "loss": 0.2513,
+      "step": 200
+    },
+    {
+      "epoch": 2.53,
+      "learning_rate": 2.0469798657718122e-05,
+      "loss": 0.2531,
+      "step": 210
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 1.8791946308724835e-05,
+      "loss": 0.2512,
+      "step": 220
+    },
+    {
+      "epoch": 2.76,
+      "learning_rate": 1.7114093959731545e-05,
+      "loss": 0.2567,
+      "step": 230
+    },
+    {
+      "epoch": 2.88,
+      "learning_rate": 1.5436241610738255e-05,
+      "loss": 0.2056,
+      "step": 240
+    },
+    {
+      "epoch": 2.99,
+      "eval_accuracy": 0.968040370058873,
+      "eval_f1": 0.968040370058873,
+      "eval_loss": 0.08735597878694534,
+      "eval_precision": 0.968040370058873,
+      "eval_recall": 0.968040370058873,
+      "eval_runtime": 73.1744,
+      "eval_samples_per_second": 16.249,
+      "eval_steps_per_second": 0.519,
+      "step": 249
+    },
+    {
+      "epoch": 3.01,
+      "learning_rate": 1.3758389261744966e-05,
+      "loss": 0.2716,
+      "step": 250
+    },
+    {
+      "epoch": 3.13,
+      "learning_rate": 1.208053691275168e-05,
+      "loss": 0.2043,
+      "step": 260
+    },
+    {
+      "epoch": 3.25,
+      "learning_rate": 1.040268456375839e-05,
+      "loss": 0.244,
+      "step": 270
+    },
+    {
+      "epoch": 3.37,
+      "learning_rate": 8.724832214765101e-06,
+      "loss": 0.1933,
+      "step": 280
+    },
+    {
+      "epoch": 3.49,
+      "learning_rate": 7.046979865771812e-06,
+      "loss": 0.2128,
+      "step": 290
+    },
+    {
+      "epoch": 3.61,
+      "learning_rate": 5.3691275167785235e-06,
+      "loss": 0.2148,
+      "step": 300
+    },
+    {
+      "epoch": 3.73,
+      "learning_rate": 3.6912751677852355e-06,
+      "loss": 0.2011,
+      "step": 310
+    },
+    {
+      "epoch": 3.85,
+      "learning_rate": 2.013422818791946e-06,
+      "loss": 0.2045,
+      "step": 320
+    },
+    {
+      "epoch": 3.97,
+      "learning_rate": 3.355704697986577e-07,
+      "loss": 0.2038,
+      "step": 330
+    },
+    {
+      "epoch": 3.99,
+      "eval_accuracy": 0.975609756097561,
+      "eval_f1": 0.975609756097561,
+      "eval_loss": 0.08477138727903366,
+      "eval_precision": 0.975609756097561,
+      "eval_recall": 0.975609756097561,
+      "eval_runtime": 63.6613,
+      "eval_samples_per_second": 18.677,
+      "eval_steps_per_second": 0.597,
+      "step": 332
+    },
+    {
+      "epoch": 3.99,
+      "step": 332,
+      "total_flos": 1.061932048026071e+18,
+      "train_loss": 0.27396771652870866,
+      "train_runtime": 7327.0516,
+      "train_samples_per_second": 5.841,
+      "train_steps_per_second": 0.045
     }
   ],
-  "max_steps": 96,
+  "max_steps": 332,
   "num_train_epochs": 4,
-  "total_flos": 3.037401854071603e+17,
+  "total_flos": 1.061932048026071e+18,
   "trial_name": null,
   "trial_params": null
 }