import gradio as gr
import numpy as np
from tensorflow.keras.models import load_model
from tensorflow.keras.preprocessing import image
import tensorflow as tf
from PIL import Image
import os

model = load_model('model.h5')

def convertToImage(file):
    with open(file, "rb") as f:
        byte = f.read(1)  # reading 1 byte at a time. returns a byte object.
        i, j, p = 0, 0, 0
        image = np.zeros((256, 256))
        temp1 = None  # Define temp1 before using it
        while byte:
            try:
                decoded_byte = byte.decode("utf-8")
                if (
                    decoded_byte == " "
                    or decoded_byte == "\r"
                    or decoded_byte == "\n"
                    or decoded_byte == "?"
                ):
                    byte = f.read(1)
                    continue
                if j > 8:
                    try:
                        image[i][p] = int(decoded_byte, 16)
                        p += 1
                    except ValueError:
                        pass
                j += 1
                if p > 255:
                    p = 0
                    j = 0
                    if i < 255:
                        i += 1
                        continue
                byte = f.read(1)
            except UnicodeDecodeError:
                byte = f.read(1)
                continue
        a = np.matrix(image)
        temp1 = a
        img = Image.fromarray(image, "L")
        img.save("a.bmp")
        
def predict_gender(img):
    pathe = os.path.splitext(img.name)
    extension = pathe[1]
    filename = pathe[0] + ".bmp"
    if(not extension == ".bmp"):
        convertToImage(img.name)
        filename = "a.bmp"
        
    img = image.load_img(filename, target_size=(256, 256))
    x = image.img_to_array(img)
    x = np.expand_dims(x, axis=0)
    x = x / 255.0

    preds = model.predict(x)

    if preds[0][0] > 0.5:
        return "Predicted malware"
    else:
        return "Predicted benign"

# Create an interface with a file input and a text output
interface = gr.Interface(
    fn=predict_gender,
    inputs=gr.inputs.File(label="Upload a file"),
    outputs="text"
)

# Launch the interface
interface.launch()