35 lines
1.1 KiB
Python
35 lines
1.1 KiB
Python
|
import tensorflow as tf
|
||
|
|
from tensorflow.keras import layers, models
|
||
|
|
import time, threading
|
||
|
|
import numpy as np
|
||
|
|
import cv2
|
||
|
|
import model_cond
|
||
|
|
|
||
|
|
noise_dim=100
|
||
|
|
|
||
|
|
generator=model_cond.generator_model()
|
||
|
|
checkpoint=tf.train.Checkpoint(generator=generator)
|
||
|
|
checkpoint.restore(tf.train.latest_checkpoint(checkpoint_dir='./training_checkpoints_gan_cond/'))
|
||
|
|
|
||
|
|
marge=20
|
||
|
|
|
||
|
|
while True:
|
||
|
|
chiffres=input("Entrez une serie de chiffre:")
|
||
|
|
try:
|
||
|
|
chiffres_int=int(chiffres)
|
||
|
|
except:
|
||
|
|
continue
|
||
|
|
|
||
|
|
liste_chiffres=[]
|
||
|
|
while (chiffres_int):
|
||
|
|
liste_chiffres.append(chiffres_int%10)
|
||
|
|
chiffres_int=int(chiffres_int/10)
|
||
|
|
seed=tf.random.normal([len(liste_chiffres), noise_dim])
|
||
|
|
labels=tf.one_hot(liste_chiffres, 10)
|
||
|
|
image=np.zeros(shape=(28+2*marge, len(liste_chiffres)*28+2*marge), dtype=np.float32)
|
||
|
|
prediction=generator([seed, labels], training=False)
|
||
|
|
for i in range(len(prediction)):
|
||
|
|
image[marge:marge+28, marge+i*28:marge+(i+1)*28]=prediction[len(liste_chiffres)-i-1, :, :, 0]*127.5+127.5
|
||
|
|
cv2.imshow("Image", image.astype(np.uint8))
|
||
|
|
key=cv2.waitKey(10)
|