Upload 7 files

app.py

@@ -0,0 +1,66 @@
+import requests
+from PIL import Image
+import gradio as gr
+from transformers import ViTImageProcessor, ViTForImageClassification
+import torch
+
+# Model and processor loading
+processor = ViTImageProcessor.from_pretrained('google/vit-base-patch16-224')
+model = ViTForImageClassification.from_pretrained('google/vit-base-patch16-224')
+
+# Function to classify image and return top 3 predictions with probabilities
+def classify_image(image):
+    try:
+        inputs = processor(images=image, return_tensors="pt")
+        outputs = model(**inputs)
+        logits = outputs.logits
+        probabilities = torch.nn.functional.softmax(logits, dim=-1)[0]
+        top3_prob, top3_catid = torch.topk(probabilities, 3)
+        
+        labels = [model.config.id2label[catid.item()] for catid in top3_catid]
+        scores = [prob.item() for prob in top3_prob]
+        
+        return {labels[i]: scores[i] for i in range(3)}
+    except Exception as e:
+        return {"Error": str(e)}
+
+INTRO_TEXT = """
+# 🎨 Zero Shot Image Classification! 🖼️
+
+<span style="font-size: 20px;">
+Upload an image and let this model predict what it is! Who knows, maybe it will surprise you! 🕵️‍♂️
+</span>
+
+Here are some links you might find interesting:
+- [GitHub](https://github.com/google-research/vision_transformer)
+- [Hugging Face Model](https://huggingface.co/google/vit-base-patch16-224)
+- [Google Research Paper](https://arxiv.org/abs/2010.11929)
+"""
+
+def create_app():
+    with gr.Blocks() as demo:
+        gr.Markdown(INTRO_TEXT)
+        
+        with gr.Row():
+            image_input = gr.Image(type="pil", label="Upload your image here!")
+            label_output = gr.Label(label="Top Predictions")
+
+        def process_image(image):
+            return classify_image(image)
+
+        image_input.change(process_image, inputs=image_input, outputs=label_output)
+        
+        gr.Markdown("## Example Images")
+        gr.Examples(
+            examples=[
+                ["cat.jpg"],
+                ["dog.jpg"],
+                ["bird.jpg"]
+            ],
+            inputs=image_input,
+        )
+
+    return demo
+
+if __name__ == "__main__":
+    create_app().launch(share=True)

deeplabv3.tflite

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ff36e24d40547fe9e645e2f4e8745d1876d6e38b332d39a82f0bf0f5d1d561b3
+size 2780176

requirements.txt

@@ -0,0 +1,9 @@
+accelerate==0.28.0
+bitsandbytes==0.43.0
+gradio==4.28.2
+scipy==1.12.0
+sentencepiece==0.2.0
+spaces==0.26.2
+torch==2.1.1
+transformers==4.40.1
+tokenizers==0.19.1

style.css

@@ -0,0 +1,53 @@
+body {
+    font-family: 'Arial', sans-serif;
+    background-color: #f9f9f9;
+    color: #333;
+    margin: 0;
+    padding: 0;
+}
+
+h1, h2, h3, p {
+    padding: 0 20px;
+}
+
+h1 {
+    text-align: center;
+    color: #4a54f1;  /* Soft blue for headers */
+    font-size: 24px;
+    margin-top: 20px;
+    margin-bottom: 10px;
+}
+
+button {
+    background-color: #4a54f1; /* Matching the header */
+    border: none;
+    color: white;
+    padding: 10px 20px;
+    text-align: center;
+    font-size: 16px;
+    border-radius: 5px;
+    cursor: pointer;
+    transition: background-color 0.3s ease;
+}
+
+button:hover {
+    background-color: #363b8c;
+}
+
+.contain {
+    max-width: 900px;
+    margin: auto;
+    padding: 1.5rem;
+    background-color: #ffffff;
+    box-shadow: 0 4px 8px rgba(0,0,0,0.1);
+    border-radius: 8px;
+}
+
+.gradio-chat {
+    border-radius: 8px;
+    box-shadow: 0 2px 5px rgba(0,0,0,0.2);
+}
+
+input[type="range"]:hover {
+    opacity: 1;
+}