trantridat/SwinBase_lora r64 alpha16 dropout0.05 batchsize64 lr0.001

README.md

@@ -19,8 +19,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [microsoft/swin-base-patch4-window7-224-in22k](https://huggingface.co/microsoft/swin-base-patch4-window7-224-in22k) on the imagefolder dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.4238
-- Accuracy: 0.8985
+- Loss: 0.4229
+- Accuracy: 0.9008
 
 ## Model description
 

all_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 99.2,
+    "eval_accuracy": 0.9007891770011274,
+    "eval_loss": 0.42286553978919983,
+    "eval_runtime": 29.6447,
+    "eval_samples_per_second": 59.842,
+    "eval_steps_per_second": 0.945,
+    "total_flos": 1.2844070249593489e+20,
+    "train_loss": 0.21447962582111357,
+    "train_runtime": 32541.0222,
+    "train_samples_per_second": 49.037,
+    "train_steps_per_second": 0.191
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 99.2,
+    "eval_accuracy": 0.9007891770011274,
+    "eval_loss": 0.42286553978919983,
+    "eval_runtime": 29.6447,
+    "eval_samples_per_second": 59.842,
+    "eval_steps_per_second": 0.945
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 99.2,
+    "total_flos": 1.2844070249593489e+20,
+    "train_loss": 0.21447962582111357,
+    "train_runtime": 32541.0222,
+    "train_samples_per_second": 49.037,
+    "train_steps_per_second": 0.191
+}