Upload app.py
Browse files
app.py
ADDED
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import gradio as gr
|
2 |
+
import tensorflow as tf
|
3 |
+
import numpy as np
|
4 |
+
from PIL import Image
|
5 |
+
|
6 |
+
#!pip install tensorflow tensorflow-datasets gradio pillow matplotlib
|
7 |
+
|
8 |
+
model_path = "Dogs-model_transferlearning.keras"
|
9 |
+
model = tf.keras.models.load_model(model_path)
|
10 |
+
|
11 |
+
# Define the core prediction function
|
12 |
+
def predict_dogs(image):
|
13 |
+
# Preprocess image
|
14 |
+
image = Image.fromarray(image.astype('uint8')) # Convert numpy array to PIL image
|
15 |
+
image = image.resize((150, 150)) # Resize the image to 150x150
|
16 |
+
image = np.array(image)
|
17 |
+
image = np.expand_dims(image, axis=0) # Add batch dimension
|
18 |
+
|
19 |
+
# Predict
|
20 |
+
prediction = model.predict(image)
|
21 |
+
|
22 |
+
# Apply softmax to get probabilities for each class
|
23 |
+
probabilities = tf.nn.softmax(prediction)
|
24 |
+
|
25 |
+
# Map probabilities to Pokemon classes
|
26 |
+
dogs_classes = ['Goldie', 'Beagle', 'Husky']
|
27 |
+
probabilities_dict = {dogs_class: round(float(probability), 2) for dogs_class, probability in zip(dogs_classes, probabilities[0])}
|
28 |
+
|
29 |
+
return probabilities_dict
|
30 |
+
|
31 |
+
# Create the Gradio interface
|
32 |
+
input_image = gr.Image()
|
33 |
+
iface = gr.Interface(
|
34 |
+
fn=predict_dogs,
|
35 |
+
inputs=input_image,
|
36 |
+
outputs=gr.Label(),
|
37 |
+
live=True,
|
38 |
+
examples=["images/01.jpg", "images/02.jpg", "images/03.jpg", "images/04.jpg", "images/06.jpg", "images/06.jpg"],
|
39 |
+
description="A simple mlp classification model for image classification using the mnist dataset.")
|
40 |
+
iface.launch()
|