Training in progress, epoch 1

all_results.json

@@ -5,9 +5,9 @@
     "eval_runtime": 0.8126,
     "eval_samples_per_second": 137.834,
     "eval_steps_per_second": 4.923,
-    "total_flos": 7.471710289148314e+16,
-    "train_loss": 0.7044070760409037,
-    "train_runtime": 101.1597,
-    "train_samples_per_second": 29.715,
-    "train_steps_per_second": 0.237
+    "total_flos": 6.644005855919309e+16,
+    "train_loss": 0.6637095979281834,
+    "train_runtime": 82.3978,
+    "train_samples_per_second": 32.44,
+    "train_steps_per_second": 0.255
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a903932b0d32682e1013a98938e515ce833c6f35a0ed8ee9f7f64c962dded96c
+oid sha256:b20dc7107761af144cebfadfbfde362e17f748462774c0dd10e32bc3d3cc7672
 size 110394865

runs/Jul24_15-46-48_52196fe4459f/events.out.tfevents.1690214188.52196fe4459f.604.3

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6d5b130d98ffa0b3594948ae44cc6218c2a31b95f6390a1a0b00fc8ad395cad5
+size 4839

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 3.0,
-    "total_flos": 7.471710289148314e+16,
-    "train_loss": 0.7044070760409037,
-    "train_runtime": 101.1597,
-    "train_samples_per_second": 29.715,
-    "train_steps_per_second": 0.237
+    "total_flos": 6.644005855919309e+16,
+    "train_loss": 0.6637095979281834,
+    "train_runtime": 82.3978,
+    "train_samples_per_second": 32.44,
+    "train_steps_per_second": 0.255
 }

trainer_state.json

@@ -1,64 +1,64 @@
 {
-  "best_metric": 0.49107142857142855,
-  "best_model_checkpoint": "swin-tiny-patch4-window7-224-jvadlamudi2/checkpoint-16",
+  "best_metric": 0.547085201793722,
+  "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": 0.44642857142857145,
-      "eval_loss": 0.7373398542404175,
-      "eval_runtime": 0.7604,
-      "eval_samples_per_second": 147.293,
-      "eval_steps_per_second": 5.26,
-      "step": 8
+      "eval_accuracy": 0.547085201793722,
+      "eval_loss": 0.7053459882736206,
+      "eval_runtime": 1.4194,
+      "eval_samples_per_second": 157.105,
+      "eval_steps_per_second": 4.932,
+      "step": 7
     },
     {
-      "epoch": 1.25,
-      "learning_rate": 3.3333333333333335e-05,
-      "loss": 0.7302,
+      "epoch": 1.43,
+      "learning_rate": 3.055555555555556e-05,
+      "loss": 0.6779,
       "step": 10
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.49107142857142855,
-      "eval_loss": 0.7152572274208069,
-      "eval_runtime": 0.7234,
-      "eval_samples_per_second": 154.816,
-      "eval_steps_per_second": 5.529,
-      "step": 16
+      "eval_accuracy": 0.5381165919282511,
+      "eval_loss": 0.6877278089523315,
+      "eval_runtime": 1.4319,
+      "eval_samples_per_second": 155.737,
+      "eval_steps_per_second": 4.889,
+      "step": 14
     },
     {
-      "epoch": 2.5,
-      "learning_rate": 9.523809523809523e-06,
-      "loss": 0.6892,
+      "epoch": 2.86,
+      "learning_rate": 2.777777777777778e-06,
+      "loss": 0.6493,
       "step": 20
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.4642857142857143,
-      "eval_loss": 0.7121878266334534,
-      "eval_runtime": 0.7337,
-      "eval_samples_per_second": 152.652,
-      "eval_steps_per_second": 5.452,
-      "step": 24
+      "eval_accuracy": 0.5426008968609866,
+      "eval_loss": 0.6919358372688293,
+      "eval_runtime": 1.8025,
+      "eval_samples_per_second": 123.715,
+      "eval_steps_per_second": 3.883,
+      "step": 21
     },
     {
       "epoch": 3.0,
-      "step": 24,
-      "total_flos": 7.471710289148314e+16,
-      "train_loss": 0.7044070760409037,
-      "train_runtime": 101.1597,
-      "train_samples_per_second": 29.715,
-      "train_steps_per_second": 0.237
+      "step": 21,
+      "total_flos": 6.644005855919309e+16,
+      "train_loss": 0.6637095979281834,
+      "train_runtime": 82.3978,
+      "train_samples_per_second": 32.44,
+      "train_steps_per_second": 0.255
     }
   ],
-  "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
 }