import cv2, glob
import tensorflow as tf
import matplotlib.pyplot as plt
import numpy as np
import os
import sqlalchemy as db
from sqlalchemy.sql import select


engine = db.create_engine('sqlite:///drawpad.db')
connection = engine.connect()
metadata = db.MetaData()
drawpad = db.Table('images',metadata,autoload=True,autoload_with = engine)
s = select([drawpad]).order_by(db.desc(drawpad.columns.img_url.label('img_url') )).limit(1)
result = connection.execute(s)
row = result.fetchone()
# print(row['img_url'])

x = row['img_url']
ImgAddress = x[15:]



print(ImgAddress)

testing ="C:/Canvas/static/dataset/" + str(ImgAddress)
print(testing)


images=glob.glob(testing)

for image in images:
    img = cv2.imread(image,0)

    re=cv2.resize(img, (int(img.shape[1]/25), int(img.shape[0]/18.75)), interpolation = cv2.INTER_AREA)

    cv2.imshow("cek",re)

    cv2.waitKey(500)
    cv2.destroyAllWindows()

    image = "testing.jpg"

    cv2.imwrite("resize/"+image, re)




CATEGORIES = ["Baju", "Celana", "Kupukupu", "Kursi", "Sepeda"]

def data(filepath):
    IMG_SIZE = 32
    img_array = cv2.imread(filepath, cv2.IMREAD_GRAYSCALE)
    new_array = cv2.resize(img_array, (IMG_SIZE, IMG_SIZE))
    plt.imshow(img_array, cmap = "gray")
    plt.show()
    return new_array.reshape(-1, IMG_SIZE, IMG_SIZE, 1)


model = tf.keras.models.load_model("maxpooling_train_data.model")
prediction = model.predict([data("C:/Canvas/resize/testing.jpg")])
J = CATEGORIES[np.argmax(prediction)]
print(J)


model = tf.keras.models.load_model("avgpooling_train_data.model")
prediction = model.predict([data("C:/Canvas/resize/testing.jpg")])
K = CATEGORIES[np.argmax(prediction)]
print(K)



coba = db.Table('Hasil',metadata,autoload=True,autoload_with = engine)
ins = coba.insert()
ins = coba.insert().values(img_url=x, hasil=J)
str(ins)
ins.compile().params
tampil=connection.execute(ins)


coba = db.Table('HasilMax',metadata,autoload=True,autoload_with = engine)
ins = coba.insert()
ins = coba.insert().values(img_url=x, hasilMax= K)
str(ins)
ins.compile().params
tampil=connection.execute(ins)