Hello,
When I do Run all cells (without any modification) in notebook C4_W4_Lab_3_CelebA_GAN_Experiments : week 4 lab 3 " Ungraded Lab: CelebA GAN Experiments" of " Generative Deep Learning with TensorFlow" course,
when I execute the cell to train the GAN and generating images :
# generate a batch of noisy input
test_z = tf.random.normal(shape=(64, 1, 1, z_dim))
# start loop
for epoch in range(30):
with tqdm(dataset) as pbar:
pbar.set_description(f"[Epoch {epoch}]")
for step, (X1, X2) in enumerate(pbar):
# train on the current batch
d_loss, g_loss, fake = train_on_batch(X1, X2)
# display the losses
pbar.set_postfix({"g_loss": g_loss.numpy(), "d_loss": d_loss.numpy()})
# generate fake images
fake_img = model_G(test_z)
# save output
if not os.path.exists(out_dir):
os.makedirs(out_dir)
file_path = out_dir+f"/epoch_{epoch:04}.png"
save_img(fake_img.numpy()[:64], file_path, 8)
# display gallery of fake faces
if epoch % 1 == 0:
with Image.open(file_path) as img:
plt.imshow(np.asarray(img))
plt.show()
At each epoch, at step 1580, I get the error :
Epoch 0]: : 1580it [01:09, 30.46it/s, g_loss=1e+4, d_loss=33.3]Exception ignored in: <function Executor.__del__ at 0x7e9554558c10>
Traceback (most recent call last):
File "/usr/local/lib/python3.10/dist-packages/tensorflow/python/eager/executor.py", line 46, in __del__
self.wait()
File "/usr/local/lib/python3.10/dist-packages/tensorflow/python/eager/executor.py", line 65, in wait
pywrap_tfe.TFE_ExecutorWaitForAllPendingNodes(self._handle)
tensorflow.python.framework.errors_impl.OutOfRangeError: End of sequence
[Epoch 0]: : 1582it [01:09, 22.84it/s, g_loss=1e+4, d_loss=33.3]
=> and the generated images does not improve and stay all yellow
Help appreciated
Best regards
Thierry