Upload 2 files

app_.py

@@ -0,0 +1,56 @@
+import gradio as gr 
+import numpy as np
+import onnxruntime as ort
+from PIL import Image 
+
+session = ort.InferenceSession(
+    "dog_emotion_classifier.onnx",
+    providers=["CPUExecutionProvider"]
+)
+
+
+def normalize_image(image):
+    
+    mean = np.array([0.485, 0.456, 0.406])
+    std = np.array([0.229, 0.224, 0.225])
+    image[0] = (image[0] - mean[0]) / std[0]
+    image[1] = (image[1] - mean[1]) / std[1]
+    image[2] = (image[2] - mean[2]) / std[2]
+    return image
+
+def preprocess_image(image, is_link = False):
+    
+    image = image.resize((288, 288),Image.BICUBIC)
+    image = np.array(image).astype(np.float32) 
+    image = normalize_image(image)
+    return image
+def predict(image_path , is_link = False):
+    image = preprocess_image(image_path , is_link)
+ 
+    image = np.reshape(image , (1, 3, 288, 288))
+    input_name = session.get_inputs()[0].name
+    outputs = session.run([session.get_outputs()[0].name], {input_name: image})
+    predictions = outputs[0]
+    probs = np.exp(predictions) / np.sum(np.exp(predictions))
+    probs = probs.flatten()
+    probs = probs.tolist()
+    to_plot  = {'angry':probs[0], 'happy':probs[1], 'relaxed':probs[2], 'sad':probs[3]}
+    
+    return to_plot
+
+
+with gr.Blocks() as demo : 
+    gr.Markdown("**Dog Emotion Classifier**")
+    with gr.Row():
+        with gr.Column():
+            gr.Markdown("Upload an image of a dog to predict its emotion.")
+            dog_image  = gr.Image(type="pil")
+            button = gr.Button("Predict")
+        
+        with gr.Column():
+            gr.Markdown("Results")
+            text_out = gr.Label()
+        value = button.click(predict, inputs=[dog_image], outputs=text_out)
+
+
+demo.launch()

requirements.txt

@@ -0,0 +1,9 @@
+gradio==5.42.0
+numpy==1.21.5
+onnxruntime==1.22.0
+pandas==2.3.1
+Pillow==9.0.1
+Pillow==11.3.0
+Requests==2.32.4
+streamlit==1.44.1
+torch==2.0.1