Training in progress, epoch 1

all_results.json

@@ -5,9 +5,9 @@
     "eval_runtime": 0.9772,
     "eval_samples_per_second": 114.611,
     "eval_steps_per_second": 4.093,
-    "total_flos": 7.471710289148314e+16,
-    "train_loss": 0.5194688042004904,
-    "train_runtime": 100.0049,
-    "train_samples_per_second": 30.059,
-    "train_steps_per_second": 0.24
+    "total_flos": 6.644005855919309e+16,
+    "train_loss": 0.5089511360440936,
+    "train_runtime": 89.722,
+    "train_samples_per_second": 29.792,
+    "train_steps_per_second": 0.234
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0489fd3e55532b8e0063a3f7ac55188574a13c18ed07abdd372e54dacd9aa68b
+oid sha256:990af9ec7d99f5a64e1c4bb762666a56f95771b419c3e7f52ebde0bdd3722f1b
 size 110394865

runs/Jul20_20-47-08_fcbe542fc008/events.out.tfevents.1689889120.fcbe542fc008.344.20

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6be690f504d72e91053eaa5d54d0c389b9b1a1dd80eab17cb6e7a497349baed7
+size 4839

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 3.0,
-    "total_flos": 7.471710289148314e+16,
-    "train_loss": 0.5194688042004904,
-    "train_runtime": 100.0049,
-    "train_samples_per_second": 30.059,
-    "train_steps_per_second": 0.24
+    "total_flos": 6.644005855919309e+16,
+    "train_loss": 0.5089511360440936,
+    "train_runtime": 89.722,
+    "train_samples_per_second": 29.792,
+    "train_steps_per_second": 0.234
 }

trainer_state.json

@@ -1,64 +1,64 @@
 {
-  "best_metric": 0.6785714285714286,
-  "best_model_checkpoint": "swin-tiny-patch4-window7-224-jvadlamudi2/checkpoint-16",
+  "best_metric": 0.7309417040358744,
+  "best_model_checkpoint": "swin-tiny-patch4-window7-224-jvadlamudi2/checkpoint-7",
   "epoch": 3.0,
-  "global_step": 24,
+  "global_step": 21,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.6517857142857143,
-      "eval_loss": 0.6572104692459106,
-      "eval_runtime": 1.399,
-      "eval_samples_per_second": 80.056,
-      "eval_steps_per_second": 2.859,
-      "step": 8
+      "eval_accuracy": 0.7309417040358744,
+      "eval_loss": 0.5390371680259705,
+      "eval_runtime": 2.0357,
+      "eval_samples_per_second": 109.545,
+      "eval_steps_per_second": 3.439,
+      "step": 7
     },
     {
-      "epoch": 1.25,
-      "learning_rate": 3.3333333333333335e-05,
-      "loss": 0.5145,
+      "epoch": 1.43,
+      "learning_rate": 3.055555555555556e-05,
+      "loss": 0.4872,
       "step": 10
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.6785714285714286,
-      "eval_loss": 0.6288004517555237,
-      "eval_runtime": 0.7518,
-      "eval_samples_per_second": 148.97,
-      "eval_steps_per_second": 5.32,
-      "step": 16
+      "eval_accuracy": 0.7040358744394619,
+      "eval_loss": 0.5610103011131287,
+      "eval_runtime": 1.5053,
+      "eval_samples_per_second": 148.139,
+      "eval_steps_per_second": 4.65,
+      "step": 14
     },
     {
-      "epoch": 2.5,
-      "learning_rate": 9.523809523809523e-06,
-      "loss": 0.5157,
+      "epoch": 2.86,
+      "learning_rate": 2.777777777777778e-06,
+      "loss": 0.5271,
       "step": 20
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.6607142857142857,
-      "eval_loss": 0.6348384618759155,
-      "eval_runtime": 0.7495,
-      "eval_samples_per_second": 149.426,
-      "eval_steps_per_second": 5.337,
-      "step": 24
+      "eval_accuracy": 0.7174887892376681,
+      "eval_loss": 0.5560548901557922,
+      "eval_runtime": 1.4756,
+      "eval_samples_per_second": 151.129,
+      "eval_steps_per_second": 4.744,
+      "step": 21
     },
     {
       "epoch": 3.0,
-      "step": 24,
-      "total_flos": 7.471710289148314e+16,
-      "train_loss": 0.5194688042004904,
-      "train_runtime": 100.0049,
-      "train_samples_per_second": 30.059,
-      "train_steps_per_second": 0.24
+      "step": 21,
+      "total_flos": 6.644005855919309e+16,
+      "train_loss": 0.5089511360440936,
+      "train_runtime": 89.722,
+      "train_samples_per_second": 29.792,
+      "train_steps_per_second": 0.234
     }
   ],
-  "max_steps": 24,
+  "max_steps": 21,
   "num_train_epochs": 3,
-  "total_flos": 7.471710289148314e+16,
+  "total_flos": 6.644005855919309e+16,
   "trial_name": null,
   "trial_params": null
 }