End of training

README.md

@@ -32,7 +32,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [microsoft/swinv2-base-patch4-window8-256](https://huggingface.co/microsoft/swinv2-base-patch4-window8-256) on the imagefolder dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6046
+- Loss: 0.5932
 - Accuracy: 0.7209
 
 ## Model description

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 9.6,
-    "eval_accuracy": 0.7441860465116279,
-    "eval_loss": 0.5859283208847046,
-    "eval_runtime": 0.3168,
-    "eval_samples_per_second": 135.736,
-    "eval_steps_per_second": 34.723,
-    "total_flos": 7.3694882123477e+16,
-    "train_loss": 0.5492916504542033,
-    "train_runtime": 34.9653,
-    "train_samples_per_second": 28.314,
-    "train_steps_per_second": 1.716
+    "eval_accuracy": 0.7209302325581395,
+    "eval_loss": 0.5932016372680664,
+    "eval_runtime": 0.4968,
+    "eval_samples_per_second": 86.55,
+    "eval_steps_per_second": 22.141,
+    "total_flos": 9.748371434294477e+16,
+    "train_loss": 0.5804096937179566,
+    "train_runtime": 57.1953,
+    "train_samples_per_second": 17.309,
+    "train_steps_per_second": 1.049
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 9.6,
-    "eval_accuracy": 0.7441860465116279,
-    "eval_loss": 0.5859283208847046,
-    "eval_runtime": 0.3168,
-    "eval_samples_per_second": 135.736,
-    "eval_steps_per_second": 34.723
+    "eval_accuracy": 0.7209302325581395,
+    "eval_loss": 0.5932016372680664,
+    "eval_runtime": 0.4968,
+    "eval_samples_per_second": 86.55,
+    "eval_steps_per_second": 22.141
 }

runs/Aug10_00-26-27_EMIMDGXA100GPU1/events.out.tfevents.1723264067.EMIMDGXA100GPU1.4104023.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ad8bb608305f1d975331400478e43662715f6b23b73697208b1785abf6ac4249
+size 405

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 9.6,
-    "total_flos": 7.3694882123477e+16,
-    "train_loss": 0.5492916504542033,
-    "train_runtime": 34.9653,
-    "train_samples_per_second": 28.314,
-    "train_steps_per_second": 1.716
+    "total_flos": 9.748371434294477e+16,
+    "train_loss": 0.5804096937179566,
+    "train_runtime": 57.1953,
+    "train_samples_per_second": 17.309,
+    "train_steps_per_second": 1.049
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.7441860465116279,
-  "best_model_checkpoint": "vit-base-patch16-224/checkpoint-6",
+  "best_metric": 0.7209302325581395,
+  "best_model_checkpoint": "vit-base-patch16-224/checkpoint-12",
   "epoch": 9.6,
   "eval_steps": 500,
   "global_step": 60,
@@ -10,144 +10,144 @@
   "log_history": [
     {
       "epoch": 0.96,
-      "eval_accuracy": 0.7441860465116279,
-      "eval_loss": 0.5859283208847046,
-      "eval_runtime": 0.2566,
-      "eval_samples_per_second": 167.599,
-      "eval_steps_per_second": 42.874,
+      "eval_accuracy": 0.3488372093023256,
+      "eval_loss": 0.742401123046875,
+      "eval_runtime": 0.4622,
+      "eval_samples_per_second": 93.03,
+      "eval_steps_per_second": 23.798,
       "step": 6
     },
     {
       "epoch": 1.6,
-      "grad_norm": 3.7175180912017822,
+      "grad_norm": 6.224728107452393,
       "learning_rate": 4.62962962962963e-06,
-      "loss": 0.605,
+      "loss": 0.7374,
       "step": 10
     },
     {
       "epoch": 1.92,
-      "eval_accuracy": 0.7441860465116279,
-      "eval_loss": 0.5841977000236511,
-      "eval_runtime": 0.2709,
-      "eval_samples_per_second": 158.721,
-      "eval_steps_per_second": 40.603,
+      "eval_accuracy": 0.7209302325581395,
+      "eval_loss": 0.5932016372680664,
+      "eval_runtime": 0.5117,
+      "eval_samples_per_second": 84.03,
+      "eval_steps_per_second": 21.496,
       "step": 12
     },
     {
       "epoch": 2.88,
-      "eval_accuracy": 0.7441860465116279,
-      "eval_loss": 0.5918598771095276,
-      "eval_runtime": 0.2627,
-      "eval_samples_per_second": 163.663,
-      "eval_steps_per_second": 41.867,
+      "eval_accuracy": 0.7209302325581395,
+      "eval_loss": 0.5843029618263245,
+      "eval_runtime": 0.5094,
+      "eval_samples_per_second": 84.414,
+      "eval_steps_per_second": 21.594,
       "step": 18
     },
     {
       "epoch": 3.2,
-      "grad_norm": 2.415996551513672,
+      "grad_norm": 6.278665065765381,
       "learning_rate": 3.7037037037037037e-06,
-      "loss": 0.5428,
+      "loss": 0.5783,
       "step": 20
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.7441860465116279,
-      "eval_loss": 0.5884882807731628,
-      "eval_runtime": 0.9618,
-      "eval_samples_per_second": 44.706,
-      "eval_steps_per_second": 11.436,
+      "eval_accuracy": 0.7209302325581395,
+      "eval_loss": 0.599581241607666,
+      "eval_runtime": 0.5003,
+      "eval_samples_per_second": 85.944,
+      "eval_steps_per_second": 21.986,
       "step": 25
     },
     {
       "epoch": 4.8,
-      "grad_norm": 2.7047529220581055,
+      "grad_norm": 12.106355667114258,
       "learning_rate": 2.7777777777777783e-06,
-      "loss": 0.5584,
+      "loss": 0.5358,
       "step": 30
     },
     {
       "epoch": 4.96,
-      "eval_accuracy": 0.7441860465116279,
-      "eval_loss": 0.588610827922821,
-      "eval_runtime": 0.2732,
-      "eval_samples_per_second": 157.373,
-      "eval_steps_per_second": 40.258,
+      "eval_accuracy": 0.7209302325581395,
+      "eval_loss": 0.6146759986877441,
+      "eval_runtime": 0.5458,
+      "eval_samples_per_second": 78.786,
+      "eval_steps_per_second": 20.154,
       "step": 31
     },
     {
       "epoch": 5.92,
-      "eval_accuracy": 0.7441860465116279,
-      "eval_loss": 0.5914657711982727,
-      "eval_runtime": 0.2503,
-      "eval_samples_per_second": 171.764,
-      "eval_steps_per_second": 43.94,
+      "eval_accuracy": 0.7209302325581395,
+      "eval_loss": 0.6159453988075256,
+      "eval_runtime": 0.4886,
+      "eval_samples_per_second": 88.012,
+      "eval_steps_per_second": 22.515,
       "step": 37
     },
     {
       "epoch": 6.4,
-      "grad_norm": 3.9282095432281494,
+      "grad_norm": 6.217154502868652,
       "learning_rate": 1.8518518518518519e-06,
-      "loss": 0.5593,
+      "loss": 0.5745,
       "step": 40
     },
     {
       "epoch": 6.88,
-      "eval_accuracy": 0.7441860465116279,
-      "eval_loss": 0.5934743881225586,
-      "eval_runtime": 0.2654,
-      "eval_samples_per_second": 162.014,
-      "eval_steps_per_second": 41.446,
+      "eval_accuracy": 0.7209302325581395,
+      "eval_loss": 0.6090837121009827,
+      "eval_runtime": 0.4608,
+      "eval_samples_per_second": 93.312,
+      "eval_steps_per_second": 23.871,
       "step": 43
     },
     {
       "epoch": 8.0,
-      "grad_norm": 2.82995343208313,
+      "grad_norm": 6.606484413146973,
       "learning_rate": 9.259259259259259e-07,
-      "loss": 0.5097,
+      "loss": 0.5325,
       "step": 50
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.7441860465116279,
-      "eval_loss": 0.5947032570838928,
-      "eval_runtime": 0.2582,
-      "eval_samples_per_second": 166.523,
-      "eval_steps_per_second": 42.599,
+      "eval_accuracy": 0.7209302325581395,
+      "eval_loss": 0.6066591739654541,
+      "eval_runtime": 0.5381,
+      "eval_samples_per_second": 79.913,
+      "eval_steps_per_second": 20.443,
       "step": 50
     },
     {
       "epoch": 8.96,
-      "eval_accuracy": 0.7441860465116279,
-      "eval_loss": 0.594927966594696,
-      "eval_runtime": 0.2847,
-      "eval_samples_per_second": 151.012,
-      "eval_steps_per_second": 38.631,
+      "eval_accuracy": 0.7209302325581395,
+      "eval_loss": 0.6046953797340393,
+      "eval_runtime": 0.5485,
+      "eval_samples_per_second": 78.392,
+      "eval_steps_per_second": 20.054,
       "step": 56
     },
     {
       "epoch": 9.6,
-      "grad_norm": 3.3591854572296143,
+      "grad_norm": 4.127104759216309,
       "learning_rate": 0.0,
-      "loss": 0.5205,
+      "loss": 0.524,
       "step": 60
     },
     {
       "epoch": 9.6,
-      "eval_accuracy": 0.7441860465116279,
-      "eval_loss": 0.5949187874794006,
-      "eval_runtime": 0.2565,
-      "eval_samples_per_second": 167.627,
-      "eval_steps_per_second": 42.881,
+      "eval_accuracy": 0.7209302325581395,
+      "eval_loss": 0.6046017408370972,
+      "eval_runtime": 0.7447,
+      "eval_samples_per_second": 57.745,
+      "eval_steps_per_second": 14.772,
       "step": 60
     },
     {
       "epoch": 9.6,
       "step": 60,
-      "total_flos": 7.3694882123477e+16,
-      "train_loss": 0.5492916504542033,
-      "train_runtime": 34.9653,
-      "train_samples_per_second": 28.314,
-      "train_steps_per_second": 1.716
+      "total_flos": 9.748371434294477e+16,
+      "train_loss": 0.5804096937179566,
+      "train_runtime": 57.1953,
+      "train_samples_per_second": 17.309,
+      "train_steps_per_second": 1.049
     }
   ],
   "logging_steps": 10,
@@ -167,7 +167,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 7.3694882123477e+16,
+  "total_flos": 9.748371434294477e+16,
   "train_batch_size": 4,
   "trial_name": null,
   "trial_params": null