Training in progress, epoch 1

all_results.json

@@ -5,9 +5,9 @@
     "eval_runtime": 0.8492,
     "eval_samples_per_second": 131.885,
     "eval_steps_per_second": 4.71,
-    "total_flos": 7.471710289148314e+16,
-    "train_loss": 0.6844675143559774,
-    "train_runtime": 103.9833,
-    "train_samples_per_second": 28.908,
-    "train_steps_per_second": 0.231
+    "total_flos": 6.644005855919309e+16,
+    "train_loss": 0.6421611678032648,
+    "train_runtime": 84.9087,
+    "train_samples_per_second": 31.481,
+    "train_steps_per_second": 0.247
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:24edaac55992451853222a124c5d8f6416b5c4ef5aaee33c8f19c4e816a01660
+oid sha256:f81bb98f862961b3a3890ff617d80e5fcd24cadcca9932bc1df02205e335ee93
 size 110394865

runs/Jul20_19-45-58_fcbe542fc008/events.out.tfevents.1689884349.fcbe542fc008.344.4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:609c55ab98547399665fea9bacb4c18f92afd38d536a71279ff8b271c0c1557a
+size 4839

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 3.0,
-    "total_flos": 7.471710289148314e+16,
-    "train_loss": 0.6844675143559774,
-    "train_runtime": 103.9833,
-    "train_samples_per_second": 28.908,
-    "train_steps_per_second": 0.231
+    "total_flos": 6.644005855919309e+16,
+    "train_loss": 0.6421611678032648,
+    "train_runtime": 84.9087,
+    "train_samples_per_second": 31.481,
+    "train_steps_per_second": 0.247
 }

trainer_state.json

@@ -1,64 +1,64 @@
 {
-  "best_metric": 0.6071428571428571,
-  "best_model_checkpoint": "swin-tiny-patch4-window7-224-jvadlamudi2/checkpoint-24",
+  "best_metric": 0.57847533632287,
+  "best_model_checkpoint": "swin-tiny-patch4-window7-224-jvadlamudi2/checkpoint-21",
   "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": 0.5892857142857143,
-      "eval_loss": 0.7051630616188049,
-      "eval_runtime": 0.9526,
-      "eval_samples_per_second": 117.567,
-      "eval_steps_per_second": 4.199,
-      "step": 8
+      "eval_accuracy": 0.5560538116591929,
+      "eval_loss": 0.6701683402061462,
+      "eval_runtime": 1.5059,
+      "eval_samples_per_second": 148.081,
+      "eval_steps_per_second": 4.648,
+      "step": 7
     },
     {
-      "epoch": 1.25,
-      "learning_rate": 3.3333333333333335e-05,
-      "loss": 0.7036,
+      "epoch": 1.43,
+      "learning_rate": 3.055555555555556e-05,
+      "loss": 0.6558,
       "step": 10
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.5803571428571429,
-      "eval_loss": 0.6873636245727539,
-      "eval_runtime": 0.7758,
-      "eval_samples_per_second": 144.369,
-      "eval_steps_per_second": 5.156,
-      "step": 16
+      "eval_accuracy": 0.5739910313901345,
+      "eval_loss": 0.6717117428779602,
+      "eval_runtime": 1.4886,
+      "eval_samples_per_second": 149.808,
+      "eval_steps_per_second": 4.702,
+      "step": 14
     },
     {
-      "epoch": 2.5,
-      "learning_rate": 9.523809523809523e-06,
-      "loss": 0.6691,
+      "epoch": 2.86,
+      "learning_rate": 2.777777777777778e-06,
+      "loss": 0.6275,
       "step": 20
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.6071428571428571,
-      "eval_loss": 0.6862359046936035,
-      "eval_runtime": 0.8007,
-      "eval_samples_per_second": 139.883,
-      "eval_steps_per_second": 4.996,
-      "step": 24
+      "eval_accuracy": 0.57847533632287,
+      "eval_loss": 0.6714264154434204,
+      "eval_runtime": 1.5058,
+      "eval_samples_per_second": 148.097,
+      "eval_steps_per_second": 4.649,
+      "step": 21
     },
     {
       "epoch": 3.0,
-      "step": 24,
-      "total_flos": 7.471710289148314e+16,
-      "train_loss": 0.6844675143559774,
-      "train_runtime": 103.9833,
-      "train_samples_per_second": 28.908,
-      "train_steps_per_second": 0.231
+      "step": 21,
+      "total_flos": 6.644005855919309e+16,
+      "train_loss": 0.6421611678032648,
+      "train_runtime": 84.9087,
+      "train_samples_per_second": 31.481,
+      "train_steps_per_second": 0.247
     }
   ],
-  "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
 }