End of training

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 20.0,
-    "eval_accuracy": 1.0,
-    "eval_loss": 0.5404492020606995,
-    "eval_runtime": 0.0787,
-    "eval_samples_per_second": 50.832,
-    "eval_steps_per_second": 12.708,
+    "eval_accuracy": 0.75,
+    "eval_loss": 0.48944157361984253,
+    "eval_runtime": 0.0856,
+    "eval_samples_per_second": 46.731,
+    "eval_steps_per_second": 11.683,
     "total_flos": 1.73991922900992e+16,
-    "train_loss": 0.15340265035629272,
-    "train_runtime": 48.7906,
-    "train_samples_per_second": 14.347,
-    "train_steps_per_second": 0.41
+    "train_loss": 0.04178592562675476,
+    "train_runtime": 88.348,
+    "train_samples_per_second": 7.923,
+    "train_steps_per_second": 0.226
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 10.0,
-    "eval_accuracy": 1.0,
-    "eval_loss": 0.5404492020606995,
-    "eval_runtime": 0.0787,
-    "eval_samples_per_second": 50.832,
-    "eval_steps_per_second": 12.708
+    "epoch": 20.0,
+    "eval_accuracy": 0.75,
+    "eval_loss": 0.48944157361984253,
+    "eval_runtime": 0.0856,
+    "eval_samples_per_second": 46.731,
+    "eval_steps_per_second": 11.683
 }

runs/Sep06_02-49-21_8a3aa9d02c1c/events.out.tfevents.1662433269.8a3aa9d02c1c.69.6

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f8eaaf960ed226c23138a38f1b5ee02294809edccd1d42bb5ecf95dac135f52
+size 357

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 20.0,
     "total_flos": 1.73991922900992e+16,
-    "train_loss": 0.15340265035629272,
-    "train_runtime": 48.7906,
-    "train_samples_per_second": 14.347,
-    "train_steps_per_second": 0.41
+    "train_loss": 0.04178592562675476,
+    "train_runtime": 88.348,
+    "train_samples_per_second": 7.923,
+    "train_steps_per_second": 0.226
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
   "best_metric": 0.75,
-  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-vosap/checkpoint-11",
+  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-vosap/checkpoint-1",
   "epoch": 20.0,
   "global_step": 20,
   "is_hyper_param_search": false,
@@ -9,204 +9,204 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.5,
-      "eval_loss": 0.6076761484146118,
-      "eval_runtime": 0.0775,
-      "eval_samples_per_second": 51.64,
-      "eval_steps_per_second": 12.91,
+      "eval_accuracy": 0.75,
+      "eval_loss": 0.48944157361984253,
+      "eval_runtime": 0.0738,
+      "eval_samples_per_second": 54.208,
+      "eval_steps_per_second": 13.552,
       "step": 1
     },
     {
       "epoch": 2.0,
       "eval_accuracy": 0.5,
-      "eval_loss": 0.5957173705101013,
-      "eval_runtime": 0.0821,
-      "eval_samples_per_second": 48.732,
-      "eval_steps_per_second": 12.183,
+      "eval_loss": 0.5364932417869568,
+      "eval_runtime": 0.073,
+      "eval_samples_per_second": 54.831,
+      "eval_steps_per_second": 13.708,
       "step": 2
     },
     {
       "epoch": 3.0,
       "eval_accuracy": 0.5,
-      "eval_loss": 0.6553595662117004,
-      "eval_runtime": 0.0814,
-      "eval_samples_per_second": 49.141,
-      "eval_steps_per_second": 12.285,
+      "eval_loss": 0.6957066059112549,
+      "eval_runtime": 0.0706,
+      "eval_samples_per_second": 56.625,
+      "eval_steps_per_second": 14.156,
       "step": 3
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.25,
-      "eval_loss": 0.7486120462417603,
-      "eval_runtime": 0.092,
-      "eval_samples_per_second": 43.462,
-      "eval_steps_per_second": 10.866,
+      "eval_accuracy": 0.5,
+      "eval_loss": 0.6780673861503601,
+      "eval_runtime": 0.073,
+      "eval_samples_per_second": 54.779,
+      "eval_steps_per_second": 13.695,
       "step": 4
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.25,
-      "eval_loss": 0.8207159042358398,
-      "eval_runtime": 0.0849,
-      "eval_samples_per_second": 47.103,
-      "eval_steps_per_second": 11.776,
+      "eval_accuracy": 0.5,
+      "eval_loss": 0.5617495179176331,
+      "eval_runtime": 0.0831,
+      "eval_samples_per_second": 48.114,
+      "eval_steps_per_second": 12.028,
       "step": 5
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.25,
-      "eval_loss": 0.8213469386100769,
-      "eval_runtime": 0.0786,
-      "eval_samples_per_second": 50.921,
-      "eval_steps_per_second": 12.73,
+      "eval_accuracy": 0.75,
+      "eval_loss": 0.4461132287979126,
+      "eval_runtime": 0.08,
+      "eval_samples_per_second": 50.009,
+      "eval_steps_per_second": 12.502,
       "step": 6
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.5,
-      "eval_loss": 0.7957496643066406,
-      "eval_runtime": 0.1037,
-      "eval_samples_per_second": 38.557,
-      "eval_steps_per_second": 9.639,
+      "eval_accuracy": 0.75,
+      "eval_loss": 0.33677467703819275,
+      "eval_runtime": 0.0791,
+      "eval_samples_per_second": 50.594,
+      "eval_steps_per_second": 12.649,
       "step": 7
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.5,
-      "eval_loss": 0.7098022699356079,
-      "eval_runtime": 0.0873,
-      "eval_samples_per_second": 45.794,
-      "eval_steps_per_second": 11.449,
+      "eval_accuracy": 0.75,
+      "eval_loss": 0.3289344906806946,
+      "eval_runtime": 0.0788,
+      "eval_samples_per_second": 50.754,
+      "eval_steps_per_second": 12.688,
       "step": 8
     },
     {
       "epoch": 9.0,
-      "eval_accuracy": 0.5,
-      "eval_loss": 0.6372093558311462,
-      "eval_runtime": 0.0953,
-      "eval_samples_per_second": 41.99,
-      "eval_steps_per_second": 10.497,
+      "eval_accuracy": 0.75,
+      "eval_loss": 0.36416471004486084,
+      "eval_runtime": 0.0784,
+      "eval_samples_per_second": 51.038,
+      "eval_steps_per_second": 12.76,
       "step": 9
     },
     {
       "epoch": 10.0,
       "learning_rate": 2.777777777777778e-05,
-      "loss": 0.2113,
+      "loss": 0.0539,
       "step": 10
     },
     {
       "epoch": 10.0,
-      "eval_accuracy": 0.5,
-      "eval_loss": 0.5358340740203857,
-      "eval_runtime": 0.085,
-      "eval_samples_per_second": 47.079,
-      "eval_steps_per_second": 11.77,
+      "eval_accuracy": 0.75,
+      "eval_loss": 0.43336328864097595,
+      "eval_runtime": 0.0779,
+      "eval_samples_per_second": 51.317,
+      "eval_steps_per_second": 12.829,
       "step": 10
     },
     {
       "epoch": 11.0,
-      "eval_accuracy": 0.75,
-      "eval_loss": 0.48944157361984253,
-      "eval_runtime": 0.0964,
-      "eval_samples_per_second": 41.48,
-      "eval_steps_per_second": 10.37,
+      "eval_accuracy": 0.5,
+      "eval_loss": 0.558241605758667,
+      "eval_runtime": 0.0837,
+      "eval_samples_per_second": 47.766,
+      "eval_steps_per_second": 11.941,
       "step": 11
     },
     {
       "epoch": 12.0,
-      "eval_accuracy": 0.75,
-      "eval_loss": 0.4506962299346924,
-      "eval_runtime": 0.111,
-      "eval_samples_per_second": 36.02,
-      "eval_steps_per_second": 9.005,
+      "eval_accuracy": 0.5,
+      "eval_loss": 0.6675510406494141,
+      "eval_runtime": 0.0841,
+      "eval_samples_per_second": 47.559,
+      "eval_steps_per_second": 11.89,
       "step": 12
     },
     {
       "epoch": 13.0,
-      "eval_accuracy": 0.75,
-      "eval_loss": 0.43113046884536743,
-      "eval_runtime": 0.0817,
-      "eval_samples_per_second": 48.933,
-      "eval_steps_per_second": 12.233,
+      "eval_accuracy": 0.5,
+      "eval_loss": 0.7585530281066895,
+      "eval_runtime": 0.0802,
+      "eval_samples_per_second": 49.85,
+      "eval_steps_per_second": 12.463,
       "step": 13
     },
     {
       "epoch": 14.0,
-      "eval_accuracy": 0.75,
-      "eval_loss": 0.4339416027069092,
-      "eval_runtime": 0.1033,
-      "eval_samples_per_second": 38.72,
-      "eval_steps_per_second": 9.68,
+      "eval_accuracy": 0.5,
+      "eval_loss": 0.7936992645263672,
+      "eval_runtime": 0.097,
+      "eval_samples_per_second": 41.223,
+      "eval_steps_per_second": 10.306,
       "step": 14
     },
     {
       "epoch": 15.0,
-      "eval_accuracy": 0.75,
-      "eval_loss": 0.46003320813179016,
-      "eval_runtime": 0.0837,
-      "eval_samples_per_second": 47.81,
-      "eval_steps_per_second": 11.952,
+      "eval_accuracy": 0.5,
+      "eval_loss": 0.7985649108886719,
+      "eval_runtime": 0.1051,
+      "eval_samples_per_second": 38.07,
+      "eval_steps_per_second": 9.518,
       "step": 15
     },
     {
       "epoch": 16.0,
       "eval_accuracy": 0.5,
-      "eval_loss": 0.49817201495170593,
-      "eval_runtime": 0.1026,
-      "eval_samples_per_second": 38.983,
-      "eval_steps_per_second": 9.746,
+      "eval_loss": 0.7618705034255981,
+      "eval_runtime": 0.1109,
+      "eval_samples_per_second": 36.061,
+      "eval_steps_per_second": 9.015,
       "step": 16
     },
     {
       "epoch": 17.0,
       "eval_accuracy": 0.5,
-      "eval_loss": 0.5298565626144409,
-      "eval_runtime": 0.0879,
-      "eval_samples_per_second": 45.53,
-      "eval_steps_per_second": 11.383,
+      "eval_loss": 0.713436484336853,
+      "eval_runtime": 0.0785,
+      "eval_samples_per_second": 50.928,
+      "eval_steps_per_second": 12.732,
       "step": 17
     },
     {
       "epoch": 18.0,
       "eval_accuracy": 0.5,
-      "eval_loss": 0.5602237582206726,
-      "eval_runtime": 0.0988,
-      "eval_samples_per_second": 40.497,
-      "eval_steps_per_second": 10.124,
+      "eval_loss": 0.6725359559059143,
+      "eval_runtime": 0.0799,
+      "eval_samples_per_second": 50.075,
+      "eval_steps_per_second": 12.519,
       "step": 18
     },
     {
       "epoch": 19.0,
       "eval_accuracy": 0.5,
-      "eval_loss": 0.5776747465133667,
-      "eval_runtime": 0.0845,
-      "eval_samples_per_second": 47.319,
-      "eval_steps_per_second": 11.83,
+      "eval_loss": 0.639014720916748,
+      "eval_runtime": 0.0779,
+      "eval_samples_per_second": 51.332,
+      "eval_steps_per_second": 12.833,
       "step": 19
     },
     {
       "epoch": 20.0,
       "learning_rate": 0.0,
-      "loss": 0.0955,
+      "loss": 0.0297,
       "step": 20
     },
     {
       "epoch": 20.0,
       "eval_accuracy": 0.5,
-      "eval_loss": 0.5813302993774414,
-      "eval_runtime": 0.101,
-      "eval_samples_per_second": 39.589,
-      "eval_steps_per_second": 9.897,
+      "eval_loss": 0.6222209334373474,
+      "eval_runtime": 0.0788,
+      "eval_samples_per_second": 50.742,
+      "eval_steps_per_second": 12.685,
       "step": 20
     },
     {
       "epoch": 20.0,
       "step": 20,
       "total_flos": 1.73991922900992e+16,
-      "train_loss": 0.15340265035629272,
-      "train_runtime": 48.7906,
-      "train_samples_per_second": 14.347,
-      "train_steps_per_second": 0.41
+      "train_loss": 0.04178592562675476,
+      "train_runtime": 88.348,
+      "train_samples_per_second": 7.923,
+      "train_steps_per_second": 0.226
     }
   ],
   "max_steps": 20,