End of training

all_results.json

@@ -1,13 +1,13 @@
 {
-    "epoch": 2.96,
-    "eval_accuracy": 0.5629370629370629,
-    "eval_loss": 0.9363093972206116,
-    "eval_runtime": 162.5089,
-    "eval_samples_per_second": 1.76,
-    "eval_steps_per_second": 0.055,
-    "total_flos": 1.8986065360883712e+17,
-    "train_loss": 1.0558061440785727,
-    "train_runtime": 7791.9181,
-    "train_samples_per_second": 0.99,
+    "epoch": 2.98,
+    "eval_accuracy": 0.5236051502145923,
+    "eval_loss": 1.0571081638336182,
+    "eval_runtime": 242.6107,
+    "eval_samples_per_second": 2.881,
+    "eval_steps_per_second": 0.091,
+    "total_flos": 4.6672213042151424e+17,
+    "train_loss": 1.1336749264983093,
+    "train_runtime": 17612.5977,
+    "train_samples_per_second": 1.07,
     "train_steps_per_second": 0.008
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 2.96,
-    "eval_accuracy": 0.5629370629370629,
-    "eval_loss": 0.9363093972206116,
-    "eval_runtime": 162.5089,
-    "eval_samples_per_second": 1.76,
-    "eval_steps_per_second": 0.055
+    "epoch": 2.98,
+    "eval_accuracy": 0.5236051502145923,
+    "eval_loss": 1.0571081638336182,
+    "eval_runtime": 242.6107,
+    "eval_samples_per_second": 2.881,
+    "eval_steps_per_second": 0.091
 }

runs/Dec17_18-23-09_04eb97d951b1/events.out.tfevents.1702855251.04eb97d951b1.2163.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:88b5eb8e90cc49c2a14447dd844dc6792ab7ab6e3a875e6b50e851687d7bec95
+size 411

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 2.96,
-    "total_flos": 1.8986065360883712e+17,
-    "train_loss": 1.0558061440785727,
-    "train_runtime": 7791.9181,
-    "train_samples_per_second": 0.99,
+    "epoch": 2.98,
+    "total_flos": 4.6672213042151424e+17,
+    "train_loss": 1.1336749264983093,
+    "train_runtime": 17612.5977,
+    "train_samples_per_second": 1.07,
     "train_steps_per_second": 0.008
 }

trainer_state.json

@@ -1,91 +1,139 @@
 {
-  "best_metric": 0.5629370629370629,
-  "best_model_checkpoint": "swin-tiny-patch4-window7-224-breastdensity/checkpoint-60",
-  "epoch": 2.962962962962963,
+  "best_metric": 0.5236051502145923,
+  "best_model_checkpoint": "swin-tiny-patch4-window7-224-breastdensity/checkpoint-147",
+  "epoch": 2.984771573604061,
   "eval_steps": 500,
-  "global_step": 60,
+  "global_step": 147,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.49,
-      "learning_rate": 4.62962962962963e-05,
-      "loss": 1.3348,
+      "epoch": 0.2,
+      "learning_rate": 3.3333333333333335e-05,
+      "loss": 1.3539,
       "step": 10
     },
     {
-      "epoch": 0.99,
-      "learning_rate": 3.7037037037037037e-05,
-      "loss": 1.1581,
+      "epoch": 0.41,
+      "learning_rate": 4.810606060606061e-05,
+      "loss": 1.2596,
       "step": 20
     },
     {
-      "epoch": 0.99,
-      "eval_accuracy": 0.46153846153846156,
-      "eval_loss": 1.1298887729644775,
-      "eval_runtime": 128.2636,
-      "eval_samples_per_second": 2.23,
-      "eval_steps_per_second": 0.07,
-      "step": 20
-    },
-    {
-      "epoch": 1.48,
-      "learning_rate": 2.777777777777778e-05,
-      "loss": 0.9956,
+      "epoch": 0.61,
+      "learning_rate": 4.431818181818182e-05,
+      "loss": 1.1889,
       "step": 30
     },
     {
-      "epoch": 1.98,
-      "learning_rate": 1.8518518518518518e-05,
-      "loss": 0.9835,
+      "epoch": 0.81,
+      "learning_rate": 4.053030303030303e-05,
+      "loss": 1.1872,
       "step": 40
     },
     {
-      "epoch": 1.98,
-      "eval_accuracy": 0.541958041958042,
-      "eval_loss": 0.9605219960212708,
-      "eval_runtime": 127.3603,
-      "eval_samples_per_second": 2.246,
-      "eval_steps_per_second": 0.071,
-      "step": 40
+      "epoch": 0.99,
+      "eval_accuracy": 0.43204577968526464,
+      "eval_loss": 1.2194271087646484,
+      "eval_runtime": 247.2453,
+      "eval_samples_per_second": 2.827,
+      "eval_steps_per_second": 0.089,
+      "step": 49
     },
     {
-      "epoch": 2.47,
-      "learning_rate": 9.259259259259259e-06,
-      "loss": 0.9577,
+      "epoch": 1.02,
+      "learning_rate": 3.6742424242424246e-05,
+      "loss": 1.1696,
       "step": 50
     },
     {
-      "epoch": 2.96,
-      "learning_rate": 0.0,
-      "loss": 0.9051,
+      "epoch": 1.22,
+      "learning_rate": 3.295454545454545e-05,
+      "loss": 1.1525,
       "step": 60
     },
     {
-      "epoch": 2.96,
-      "eval_accuracy": 0.5629370629370629,
-      "eval_loss": 0.9363093972206116,
-      "eval_runtime": 124.1301,
-      "eval_samples_per_second": 2.304,
-      "eval_steps_per_second": 0.073,
-      "step": 60
+      "epoch": 1.42,
+      "learning_rate": 2.916666666666667e-05,
+      "loss": 1.1006,
+      "step": 70
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 2.537878787878788e-05,
+      "loss": 1.0961,
+      "step": 80
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 2.1590909090909093e-05,
+      "loss": 1.0998,
+      "step": 90
+    },
+    {
+      "epoch": 1.99,
+      "eval_accuracy": 0.48068669527896996,
+      "eval_loss": 1.0917412042617798,
+      "eval_runtime": 254.7856,
+      "eval_samples_per_second": 2.743,
+      "eval_steps_per_second": 0.086,
+      "step": 98
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 1.7803030303030303e-05,
+      "loss": 1.0582,
+      "step": 100
+    },
+    {
+      "epoch": 2.23,
+      "learning_rate": 1.4015151515151515e-05,
+      "loss": 1.0765,
+      "step": 110
+    },
+    {
+      "epoch": 2.44,
+      "learning_rate": 1.0227272727272729e-05,
+      "loss": 1.0685,
+      "step": 120
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 6.43939393939394e-06,
+      "loss": 1.0714,
+      "step": 130
+    },
+    {
+      "epoch": 2.84,
+      "learning_rate": 2.651515151515152e-06,
+      "loss": 1.0623,
+      "step": 140
+    },
+    {
+      "epoch": 2.98,
+      "eval_accuracy": 0.5236051502145923,
+      "eval_loss": 1.0571081638336182,
+      "eval_runtime": 249.3354,
+      "eval_samples_per_second": 2.803,
+      "eval_steps_per_second": 0.088,
+      "step": 147
     },
     {
-      "epoch": 2.96,
-      "step": 60,
-      "total_flos": 1.8986065360883712e+17,
-      "train_loss": 1.0558061440785727,
-      "train_runtime": 7791.9181,
-      "train_samples_per_second": 0.99,
+      "epoch": 2.98,
+      "step": 147,
+      "total_flos": 4.6672213042151424e+17,
+      "train_loss": 1.1336749264983093,
+      "train_runtime": 17612.5977,
+      "train_samples_per_second": 1.07,
       "train_steps_per_second": 0.008
     }
   ],
   "logging_steps": 10,
-  "max_steps": 60,
+  "max_steps": 147,
   "num_train_epochs": 3,
   "save_steps": 500,
-  "total_flos": 1.8986065360883712e+17,
+  "total_flos": 4.6672213042151424e+17,
   "trial_name": null,
   "trial_params": null
 }