Training in progress, epoch 0

all_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 2.88,
-    "total_flos": 5.727299925757133e+16,
-    "train_loss": 1.1426812807718914,
-    "train_runtime": 417.2641,
-    "train_samples_per_second": 5.752,
-    "train_steps_per_second": 0.043
+    "total_flos": 1.1454599851514266e+17,
+    "train_loss": 1.034223583009508,
+    "train_runtime": 597.4423,
+    "train_samples_per_second": 8.034,
+    "train_steps_per_second": 0.06
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2754a8794b28d4f477531124bf80b3acd335749104df0ab90e4530c0be87d992
+oid sha256:873780e294ef91fadba6ed031eaa71831cac484c699d9fc97b30c7e7a1644225
 size 110352060

runs/Aug19_00-04-27_e8f7099b712a/events.out.tfevents.1724026490.e8f7099b712a.297.6

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:db92d3350eeb0bfbc07ae235a95a047573e5fd3788e7e20d8b3bea3beb3e7037
+size 404

runs/Aug19_00-20-42_e8f7099b712a/events.out.tfevents.1724026848.e8f7099b712a.297.7

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3e83c9f698580c30e14d8710a7b3546e16687d9d064198ca4258142c4a8e658e
+size 6292

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 2.88,
-    "total_flos": 5.727299925757133e+16,
-    "train_loss": 1.1426812807718914,
-    "train_runtime": 417.2641,
-    "train_samples_per_second": 5.752,
-    "train_steps_per_second": 0.043
+    "total_flos": 1.1454599851514266e+17,
+    "train_loss": 1.034223583009508,
+    "train_runtime": 597.4423,
+    "train_samples_per_second": 8.034,
+    "train_steps_per_second": 0.06
 }

trainer_state.json

@@ -1,68 +1,82 @@
 {
-  "best_metric": 1.0755484104156494,
-  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned/checkpoint-18",
+  "best_metric": 0.871762752532959,
+  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned/checkpoint-25",
   "epoch": 2.88,
   "eval_steps": 500,
-  "global_step": 18,
+  "global_step": 36,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
+    {
+      "epoch": 0.8,
+      "grad_norm": 5.137752532958984,
+      "learning_rate": 4.0625000000000005e-05,
+      "loss": 1.2933,
+      "step": 10
+    },
     {
       "epoch": 0.96,
       "eval_accuracy": {
-        "accuracy": 0.415
+        "accuracy": 0.5525
       },
-      "eval_logLoss": 1.196082353591919,
-      "eval_loss": 1.196082592010498,
-      "eval_runtime": 68.3725,
-      "eval_samples_per_second": 2.925,
-      "eval_steps_per_second": 0.102,
-      "step": 6
+      "eval_logLoss": 1.009933590888977,
+      "eval_loss": 1.0099334716796875,
+      "eval_runtime": 92.538,
+      "eval_samples_per_second": 4.323,
+      "eval_steps_per_second": 0.14,
+      "step": 12
     },
     {
       "epoch": 1.6,
-      "grad_norm": 3.4481194019317627,
+      "grad_norm": 4.288869857788086,
       "learning_rate": 2.5e-05,
-      "loss": 1.2502,
-      "step": 10
+      "loss": 0.9637,
+      "step": 20
     },
     {
-      "epoch": 1.92,
+      "epoch": 2.0,
       "eval_accuracy": {
-        "accuracy": 0.49
+        "accuracy": 0.6125
       },
-      "eval_logLoss": 1.0786902904510498,
-      "eval_loss": 1.0786901712417603,
-      "eval_runtime": 4.6591,
-      "eval_samples_per_second": 42.927,
-      "eval_steps_per_second": 1.502,
-      "step": 12
+      "eval_logLoss": 0.871762752532959,
+      "eval_loss": 0.871762752532959,
+      "eval_runtime": 9.6171,
+      "eval_samples_per_second": 41.593,
+      "eval_steps_per_second": 1.352,
+      "step": 25
+    },
+    {
+      "epoch": 2.4,
+      "grad_norm": 6.406833171844482,
+      "learning_rate": 9.375000000000001e-06,
+      "loss": 0.9461,
+      "step": 30
     },
     {
       "epoch": 2.88,
       "eval_accuracy": {
-        "accuracy": 0.48
+        "accuracy": 0.6225
       },
-      "eval_logLoss": 1.0755484104156494,
-      "eval_loss": 1.0755484104156494,
-      "eval_runtime": 4.8873,
-      "eval_samples_per_second": 40.922,
-      "eval_steps_per_second": 1.432,
-      "step": 18
+      "eval_logLoss": 0.8781986236572266,
+      "eval_loss": 0.8781986832618713,
+      "eval_runtime": 10.2528,
+      "eval_samples_per_second": 39.014,
+      "eval_steps_per_second": 1.268,
+      "step": 36
     },
     {
       "epoch": 2.88,
-      "step": 18,
-      "total_flos": 5.727299925757133e+16,
-      "train_loss": 1.1426812807718914,
-      "train_runtime": 417.2641,
-      "train_samples_per_second": 5.752,
-      "train_steps_per_second": 0.043
+      "step": 36,
+      "total_flos": 1.1454599851514266e+17,
+      "train_loss": 1.034223583009508,
+      "train_runtime": 597.4423,
+      "train_samples_per_second": 8.034,
+      "train_steps_per_second": 0.06
     }
   ],
   "logging_steps": 10,
-  "max_steps": 18,
+  "max_steps": 36,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 3,
   "save_steps": 500,
@@ -78,7 +92,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 5.727299925757133e+16,
+  "total_flos": 1.1454599851514266e+17,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": null

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:776601c03cf5d73011085c6726d5efd93d6819be3f8baa0b4540e1fb6b0763b7
+oid sha256:9f0825e68bcb0186ea443bdd539e6b3ef6ceb12ea12325f938bafef7df257c92
 size 5240