Training in progress, epoch 1

all_results.json

@@ -5,9 +5,9 @@
     "eval_runtime": 0.9902,
     "eval_samples_per_second": 113.112,
     "eval_steps_per_second": 4.04,
-    "total_flos": 7.471710289148314e+16,
-    "train_loss": 0.6960849761962891,
-    "train_runtime": 113.223,
-    "train_samples_per_second": 26.549,
-    "train_steps_per_second": 0.212
+    "total_flos": 6.644005855919309e+16,
+    "train_loss": 0.6436980593772161,
+    "train_runtime": 88.5395,
+    "train_samples_per_second": 30.19,
+    "train_steps_per_second": 0.237
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bfc3e33cfa055c99e19152bb2a014cfee6a65e1d15d4c354d69777716a44d413
+oid sha256:5f9f2af55cf6d79c53e4f56939600f7cde6664b6e0ce04a3884226c9d241475d
 size 110394865

runs/Jul20_20-47-08_fcbe542fc008/events.out.tfevents.1689886992.fcbe542fc008.344.12

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a3f97bb04147661ade8212252b28c9a4bee1480841191634123ed62e40c835f2
+size 4839

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 3.0,
-    "total_flos": 7.471710289148314e+16,
-    "train_loss": 0.6960849761962891,
-    "train_runtime": 113.223,
-    "train_samples_per_second": 26.549,
-    "train_steps_per_second": 0.212
+    "total_flos": 6.644005855919309e+16,
+    "train_loss": 0.6436980593772161,
+    "train_runtime": 88.5395,
+    "train_samples_per_second": 30.19,
+    "train_steps_per_second": 0.237
 }

trainer_state.json

@@ -1,64 +1,64 @@
 {
-  "best_metric": 0.5357142857142857,
-  "best_model_checkpoint": "swin-tiny-patch4-window7-224-jvadlamudi2/checkpoint-24",
+  "best_metric": 0.5964125560538116,
+  "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": 0.5178571428571429,
-      "eval_loss": 0.7408849596977234,
-      "eval_runtime": 1.0843,
-      "eval_samples_per_second": 103.293,
-      "eval_steps_per_second": 3.689,
-      "step": 8
+      "eval_accuracy": 0.5964125560538116,
+      "eval_loss": 0.6732161045074463,
+      "eval_runtime": 1.4803,
+      "eval_samples_per_second": 150.647,
+      "eval_steps_per_second": 4.729,
+      "step": 7
     },
     {
-      "epoch": 1.25,
-      "learning_rate": 3.3333333333333335e-05,
-      "loss": 0.7308,
+      "epoch": 1.43,
+      "learning_rate": 3.055555555555556e-05,
+      "loss": 0.658,
       "step": 10
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.5178571428571429,
-      "eval_loss": 0.6999436616897583,
-      "eval_runtime": 1.2067,
-      "eval_samples_per_second": 92.814,
-      "eval_steps_per_second": 3.315,
-      "step": 16
+      "eval_accuracy": 0.5874439461883408,
+      "eval_loss": 0.6781052947044373,
+      "eval_runtime": 1.4909,
+      "eval_samples_per_second": 149.572,
+      "eval_steps_per_second": 4.695,
+      "step": 14
     },
     {
-      "epoch": 2.5,
-      "learning_rate": 9.523809523809523e-06,
-      "loss": 0.6691,
+      "epoch": 2.86,
+      "learning_rate": 2.777777777777778e-06,
+      "loss": 0.6276,
       "step": 20
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.5357142857142857,
-      "eval_loss": 0.7016571164131165,
-      "eval_runtime": 1.019,
-      "eval_samples_per_second": 109.907,
-      "eval_steps_per_second": 3.925,
-      "step": 24
+      "eval_accuracy": 0.57847533632287,
+      "eval_loss": 0.6808013916015625,
+      "eval_runtime": 1.5079,
+      "eval_samples_per_second": 147.89,
+      "eval_steps_per_second": 4.642,
+      "step": 21
     },
     {
       "epoch": 3.0,
-      "step": 24,
-      "total_flos": 7.471710289148314e+16,
-      "train_loss": 0.6960849761962891,
-      "train_runtime": 113.223,
-      "train_samples_per_second": 26.549,
-      "train_steps_per_second": 0.212
+      "step": 21,
+      "total_flos": 6.644005855919309e+16,
+      "train_loss": 0.6436980593772161,
+      "train_runtime": 88.5395,
+      "train_samples_per_second": 30.19,
+      "train_steps_per_second": 0.237
     }
   ],
-  "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
 }