🔧 fix dataset

README.md

@@ -15,7 +15,7 @@ model-index:
       name: Image Classification
       type: image-classification
     dataset:
-      name: beans
+      name: customize
       type: imagefolder
       config: default
       split: validation
@@ -31,9 +31,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # vit-base-id-card
 
-This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the beans dataset.
+This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the customize dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0611
+- Loss: 0.0594
 - Accuracy: 1.0
 
 ## Model description
@@ -60,7 +60,6 @@ The following hyperparameters were used during training:
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - num_epochs: 4
-- mixed_precision_training: Native AMP
 
 ### Training results
 
@@ -69,6 +68,6 @@ The following hyperparameters were used during training:
 ### Framework versions
 
 - Transformers 4.38.2
-- Pytorch 2.2.1+cu121
+- Pytorch 2.2.1
 - Datasets 2.18.0
 - Tokenizers 0.15.2

all_results.json

@@ -1,13 +1,12 @@
 {
     "epoch": 4.0,
     "eval_accuracy": 1.0,
-    "eval_loss": 0.0610995814204216,
-    "eval_runtime": 1.8578,
-    "eval_samples_per_second": 14.533,
-    "eval_steps_per_second": 2.153,
-    "total_flos": 3.502669323615437e+16,
-    "train_loss": 0.25591564178466797,
-    "train_runtime": 40.8688,
-    "train_samples_per_second": 11.06,
-    "train_steps_per_second": 0.783
+    "eval_loss": 0.05941523611545563,
+    "eval_runtime": 2.2052,
+    "eval_samples_per_second": 12.244,
+    "eval_steps_per_second": 1.814,
+    "train_loss": 0.25243260501883924,
+    "train_runtime": 85.4792,
+    "train_samples_per_second": 5.288,
+    "train_steps_per_second": 0.374
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 4.0,
     "eval_accuracy": 1.0,
-    "eval_loss": 0.0610995814204216,
-    "eval_runtime": 1.8578,
-    "eval_samples_per_second": 14.533,
-    "eval_steps_per_second": 2.153
+    "eval_loss": 0.05941523611545563,
+    "eval_runtime": 2.2052,
+    "eval_samples_per_second": 12.244,
+    "eval_steps_per_second": 1.814
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:17fddba4b9b119ad3ef2426885782d2b79033eb714640f3d62da164f57c27206
+oid sha256:bff36a0971def28059fa5de42fe3d66e08a9c955e842c930f175761510dbb14f
 size 343227052

preprocessor_config.json

@@ -7,7 +7,7 @@
     0.5,
     0.5
   ],
-  "image_processor_type": "ViTFeatureExtractor",
+  "image_processor_type": "ViTImageProcessor",
   "image_std": [
     0.5,
     0.5,

train_results.json

@@ -1,8 +1,7 @@
 {
     "epoch": 4.0,
-    "total_flos": 3.502669323615437e+16,
-    "train_loss": 0.25591564178466797,
-    "train_runtime": 40.8688,
-    "train_samples_per_second": 11.06,
-    "train_steps_per_second": 0.783
+    "train_loss": 0.25243260501883924,
+    "train_runtime": 85.4792,
+    "train_samples_per_second": 5.288,
+    "train_steps_per_second": 0.374
 }

trainer_state.json

@@ -10,33 +10,33 @@
   "log_history": [
     {
       "epoch": 1.25,
-      "grad_norm": 1.040573000907898,
-      "learning_rate": 0.00014375,
-      "loss": 0.6083,
+      "grad_norm": 0.9073283672332764,
+      "learning_rate": 0.0001375,
+      "loss": 0.6222,
       "step": 10
     },
     {
       "epoch": 2.5,
-      "grad_norm": 0.4061480462551117,
-      "learning_rate": 8.125000000000001e-05,
-      "loss": 0.1341,
+      "grad_norm": 0.34812086820602417,
+      "learning_rate": 7.500000000000001e-05,
+      "loss": 0.1155,
       "step": 20
     },
     {
       "epoch": 3.75,
-      "grad_norm": 0.30543988943099976,
-      "learning_rate": 1.8750000000000002e-05,
-      "loss": 0.0648,
+      "grad_norm": 0.27614542841911316,
+      "learning_rate": 1.25e-05,
+      "loss": 0.0591,
       "step": 30
     },
     {
       "epoch": 4.0,
       "step": 32,
       "total_flos": 3.502669323615437e+16,
-      "train_loss": 0.25591564178466797,
-      "train_runtime": 40.8688,
-      "train_samples_per_second": 11.06,
-      "train_steps_per_second": 0.783
+      "train_loss": 0.25243260501883924,
+      "train_runtime": 85.4792,
+      "train_samples_per_second": 5.288,
+      "train_steps_per_second": 0.374
     }
   ],
   "logging_steps": 10,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ef34432b313669942c77076b60d487bda6024f60f7696c7295a2118b4ca0d8b5
-size 4856
+oid sha256:b30cbce2611a7d439fc79c26cb16d49c03b291372677674fc44afb3d2c9498cd
+size 4920