use pipeline

.gitignore

@@ -0,0 +1,2 @@
+/env/
+/__pycache__/

app.py

@@ -1,21 +1,17 @@
-from transformers import ViTImageProcessor, ViTForImageClassification
+from transformers import pipeline
 from PIL import Image
-import requests
 import os 
 import gradio as gr
 from timeit import default_timer as timer
 from typing import Tuple, Dict
 
-
 def predict(img) -> Tuple[Dict, float]:
     start_time = timer()
-    processor = ViTImageProcessor.from_pretrained('bazyl/gtsrb-model')
-    model = ViTForImageClassification.from_pretrained('bazyl/gtsrb-model')
-    inputs = processor(images=img, return_tensors="pt")
-    outputs = model(**inputs)
-    logits = outputs.logits
-    predicted_class_idx = logits.argmax(-1).item()
-    print("Predicted class:", model.config.id2label[predicted_class_idx])
+    classifier = pipeline("image-classification", model="bazyl/gtsrb-model", tokenizer="bazyl/gtsrb-model")
+    result = classifier(img)
+    response = {result[i]["label"]: result[i]["score"] for i in range(len(result))}
+    pred_time = round(timer() - start_time, 5)
+    return response, pred_time
 
 title = "GTSRB - German Traffic Sign Recognition by Bazyl Horsey"
 description = "CNN created for the GTSRB Dataset, achieved 99.93% test accuracy"

requirements.txt

@@ -1,5 +1,3 @@
 gradio==3.28.3
 Pillow==9.5.0
-Requests==2.30.0
-transformers==4.28.1
-torch==2.0.0
+transformers==4.28.1

test.py

@@ -0,0 +1,22 @@
+from transformers import ViTImageProcessor, ViTForImageClassification, AutoModelForImageClassification, AutoTokenizer, pipeline
+
+from PIL import Image
+import requests
+import os 
+import gradio as gr
+from timeit import default_timer as timer
+from typing import Tuple, Dict
+import torch
+
+start_time = timer()
+# config = AutoConfig.from_pretrained('bazyl/gtsrb-model')
+
+
+image = Image.open('examples/00009.png')
+classifier = pipeline("image-classification", model="bazyl/gtsrb-model", tokenizer="bazyl/gtsrb-model")
+result = classifier(image)
+response = {result[i]["label"]: result[i]["score"] for i in range(len(result))}
+# Calculate the prediction time
+pred_time = round(timer() - start_time, 5)
+
+print(classifier(image), pred_time)