# Trained on 80 epochs at Batch size 256 with Learning Rate of 0.001 under 50 seconds!!
from PIL import Image, ImageOps
import numpy as np
import pandas as pd
import streamlit as st
import tensorflow as tf

# Suppress warnings
import warnings
warnings.filterwarnings('ignore', category=UserWarning)
warnings.filterwarnings('ignore', category=FutureWarning)

# Disable scientific notation for clarity
np.set_printoptions(suppress=True)

def load_model_teachable():
    # Load the model
    model = tf.keras.models.load_model("cnn_model.h5")
    return model

with st.spinner('Model is being loaded..'):
    model = load_model_teachable()

st.write("""
         # Teachable Machine Model 
         """
         )

st.sidebar.info("You should be happy if it classifies as - Healthy plant 🙂")

file = st.file_uploader("Upload the image to be classified", type=["jpg", "png"])
st.set_option('deprecation.showfileUploaderEncoding', False)


def upload_predict_teachable(image, model, class_names):

    # Create the array of the right shape to feed into the keras model
    # The 'length' or number of images you can put into the array is
    # determined by the first position in the shape tuple, in this case 1
    data = np.ndarray(shape=(1, 224, 224, 3), dtype=np.float32)

    # Replace this with the path to your image
    image_RGB = image.convert("RGB")

    # resizing the image to be at least 224x224 and then cropping from the center
    image_resized = image_RGB.resize((224, 224), resample=Image.LANCZOS)

    # turn the image into a numpy array
    image_array = np.asarray(image_resized)

    # Normalize the image
    normalized_image_array = (image_array.astype(np.float32) / 127.5) - 1

    # Load the image into the array
    data[0] = normalized_image_array

    # Predicts the model
    prediction = model.predict(data)
    return prediction


if file is None:
    st.text("Please upload an image file")
else:
    image = Image.open(file)
    st.image(image, use_column_width=True)
    
    # Load the labels
    class_names = open("pages/labels.txt", "r").readlines()
    prediction = upload_predict_teachable(image, model,class_names)
    index = np.argmax(prediction)
    class_name = class_names[index]
    confidence_score = prediction[0][index]

    # Print prediction and confidence score
    result = f"Your plant is suffering from: {class_name[2:]}"
    result_score = "✅ Accurate prediction score is: {} / 100".format("%.2f"  %  confidence_score)
    st.success(result)
    st.info(result_score)


st.markdown("""
    #### Know more about this disease and suggestions to prevent them from spreading.

    - *Just copy the disease name (result) from above & paste it in the below input box as it is!*
""")

# Loading diseases info & management strategies
df = pd.read_csv('pages/Disease_solutions.csv')
disease = st.text_input('Enter the disease name below 👇', '')

# Selecting rows based on condition
about = df.loc[df['Diseases'] == disease, ['Type', 'Description']]
strategies = df.loc[df['Diseases'] == disease, 'Management Strategies']

# Print diseases' other info
tab1, tab2 = st.tabs(["👀 Description", "⭐ Solution"])

with tab1:
    st.dataframe(about, use_container_width=True)
        
with tab2:
    st.dataframe(strategies, use_container_width=True)