import streamlit as st
from PIL import Image
import numpy as np
import tensorflow as tf
from tensorflow import keras
import matplotlib.pyplot as plt
import tensorflow_hub as hub
hide_streamlit_style = """
"""
st.markdown(hide_streamlit_style, unsafe_allow_html = True)
st.title('Plant Disease Prediction')
st.write("This model is capable of predicting plant disease as a demo")
def main() :
file_uploaded = st.file_uploader('Choose an image...', type = 'jpg')
if file_uploaded is not None :
image = Image.open(file_uploaded)
st.write("Uploaded Image.")
figure = plt.figure()
plt.imshow(image)
plt.axis('off')
st.pyplot(figure)
result, confidence = predict_class(image)
st.write('Prediction : {}'.format(result))
st.write('Confidence : {}%'.format(confidence))
def predict_class(image) :
with st.spinner('Loading Model...'):
classifier_model = keras.models.load_model(r'plant_badr_model.h5', compile = False)
shape = ((200,200,3))
model = keras.Sequential([hub.KerasLayer(classifier_model, input_shape = shape)]) # ye bhi kaam kar raha he
test_image = image.resize((200, 200))
test_image = keras.preprocessing.image.img_to_array(test_image)
test_image /= 256.0
test_image = np.expand_dims(test_image, axis = 0)
class_name = list(range(0, 37))
prediction = model.predict_generator(test_image)
confidence = round(100 * (np.max(prediction[0])), 2)
final_pred = class_name[np.argmax(prediction)]
return final_pred, confidence
footer = """
"""
st.markdown(footer, unsafe_allow_html = True)
if __name__ == "__main__":
main()