🍻 test2

all_results.json

@@ -1,12 +1,12 @@
 {
     "epoch": 50.0,
-    "eval_accuracy": 0.5,
-    "eval_loss": 0.9220008850097656,
-    "eval_runtime": 0.1182,
-    "eval_samples_per_second": 33.838,
-    "eval_steps_per_second": 8.459,
-    "train_loss": 0.027269574254751204,
-    "train_runtime": 19.4285,
-    "train_samples_per_second": 41.177,
-    "train_steps_per_second": 2.574
+    "eval_accuracy": 0.75,
+    "eval_loss": 0.7729406356811523,
+    "eval_runtime": 0.1206,
+    "eval_samples_per_second": 33.163,
+    "eval_steps_per_second": 8.291,
+    "train_loss": 0.04134651497006416,
+    "train_runtime": 38.0787,
+    "train_samples_per_second": 26.261,
+    "train_steps_per_second": 2.626
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 50.0,
-    "eval_accuracy": 0.5,
-    "eval_loss": 0.9220008850097656,
-    "eval_runtime": 0.1182,
-    "eval_samples_per_second": 33.838,
-    "eval_steps_per_second": 8.459
+    "eval_accuracy": 0.75,
+    "eval_loss": 0.7729406356811523,
+    "eval_runtime": 0.1206,
+    "eval_samples_per_second": 33.163,
+    "eval_steps_per_second": 8.291
 }

runs/Aug16_09-40-09_surveily-ag-02/events.out.tfevents.1692171866.surveily-ag-02.3991.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:02d84a061b71ce2547acd36fe72554f30e797292418e3c5182cbf686bb32efb2
+size 405

train_results.json

@@ -1,7 +1,7 @@
 {
     "epoch": 50.0,
-    "train_loss": 0.027269574254751204,
-    "train_runtime": 19.4285,
-    "train_samples_per_second": 41.177,
-    "train_steps_per_second": 2.574
+    "train_loss": 0.04134651497006416,
+    "train_runtime": 38.0787,
+    "train_samples_per_second": 26.261,
+    "train_steps_per_second": 2.626
 }

trainer_state.json

@@ -1,55 +1,94 @@
 {
-  "best_metric": null,
-  "best_model_checkpoint": null,
+  "best_metric": 0.7729406356811523,
+  "best_model_checkpoint": "./vit-base/checkpoint-100",
   "epoch": 50.0,
-  "global_step": 50,
+  "global_step": 100,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 10.0,
-      "learning_rate": 0.00016,
-      "loss": 0.1359,
+      "epoch": 5.0,
+      "learning_rate": 0.000182,
+      "loss": 0.4121,
       "step": 10
     },
     {
-      "epoch": 20.0,
-      "learning_rate": 0.00012,
-      "loss": 0.0002,
+      "epoch": 10.0,
+      "learning_rate": 0.000162,
+      "loss": 0.0007,
       "step": 20
     },
     {
-      "epoch": 30.0,
-      "learning_rate": 8e-05,
-      "loss": 0.0001,
+      "epoch": 15.0,
+      "learning_rate": 0.000142,
+      "loss": 0.0002,
       "step": 30
     },
     {
-      "epoch": 40.0,
-      "learning_rate": 4e-05,
+      "epoch": 20.0,
+      "learning_rate": 0.000122,
       "loss": 0.0001,
       "step": 40
     },
     {
-      "epoch": 50.0,
-      "learning_rate": 0.0,
+      "epoch": 25.0,
+      "learning_rate": 0.00010200000000000001,
       "loss": 0.0001,
       "step": 50
     },
+    {
+      "epoch": 30.0,
+      "learning_rate": 8.2e-05,
+      "loss": 0.0001,
+      "step": 60
+    },
+    {
+      "epoch": 35.0,
+      "learning_rate": 6.2e-05,
+      "loss": 0.0001,
+      "step": 70
+    },
+    {
+      "epoch": 40.0,
+      "learning_rate": 4.2e-05,
+      "loss": 0.0001,
+      "step": 80
+    },
+    {
+      "epoch": 45.0,
+      "learning_rate": 2.2000000000000003e-05,
+      "loss": 0.0001,
+      "step": 90
+    },
+    {
+      "epoch": 50.0,
+      "learning_rate": 2.0000000000000003e-06,
+      "loss": 0.0,
+      "step": 100
+    },
+    {
+      "epoch": 50.0,
+      "eval_accuracy": 0.75,
+      "eval_loss": 0.7729406356811523,
+      "eval_runtime": 0.1172,
+      "eval_samples_per_second": 34.137,
+      "eval_steps_per_second": 8.534,
+      "step": 100
+    },
     {
       "epoch": 50.0,
-      "step": 50,
-      "total_flos": 6.19947029495808e+16,
-      "train_loss": 0.027269574254751204,
-      "train_runtime": 19.4285,
-      "train_samples_per_second": 41.177,
-      "train_steps_per_second": 2.574
+      "step": 100,
+      "total_flos": 7.7493378686976e+16,
+      "train_loss": 0.04134651497006416,
+      "train_runtime": 38.0787,
+      "train_samples_per_second": 26.261,
+      "train_steps_per_second": 2.626
     }
   ],
-  "max_steps": 50,
+  "max_steps": 100,
   "num_train_epochs": 50,
-  "total_flos": 6.19947029495808e+16,
+  "total_flos": 7.7493378686976e+16,
   "trial_name": null,
   "trial_params": null
 }