tanishqgautam / image-captioning Goto Github PK

Hey @jelifysh ,

Great implementation. Thanks a lot.
Is there an implementation with beam search for predicting the caption? Will be much helpful.

Hello @jelifysh ,

Nice program, I got some error at this step, please try to resolve this
error is:
module 'tensorflow._api.v1.data.experimental' has no attribute 'AUTOTUNE'

in this step:
image_dataset = image_dataset.map(
load_image, num_parallel_calls=tf.data.experimental.AUTOTUNE).batch(16)

for epoch in range(30):
start = time.time()

train_loss.reset_states()
train_accuracy.reset_states()

for (batch, (img_tensor, tar)) in enumerate(dataset):
train_step(img_tensor, tar)

if batch % 50 == 0:
  print ('Epoch {} Batch {} Loss {:.4f} Accuracy {:.4f}'.format(
      epoch + 1, batch, train_loss.result(), train_accuracy.result()))

print ('Epoch {} Loss {:.4f} Accuracy {:.4f}'.format(epoch + 1,
train_loss.result(),
train_accuracy.result()))

print ('Time taken for 1 epoch: {} secs\n'.format(time.time() - start))

Above same code of yours, I am using after running for maximum 9 epochs its simply loading not going further even waiting for an hour its just loading Please help me out to resolve this issue as soon as passible.
code link-https://colab.research.google.com/drive/1PnMF36tgjT0vhnGNJBbu4qW1Xz99d3O0#scrollTo=vUWj0zWV4ePj

Thank you

Dear Bro,
hope you doing well. i read your code, it very helpful for me.
But could you please suggest me how to save weight of the model after train? and then. when we want predict, we only load weight ( not to trainning again):

link to the code is below:
https://github.com/tanishqgautam/Image-Captioning/blob/main/Attention/Image_caption_using_attention.ipynb

Thank in advance!

error in
for img, path in tqdm(image_dataset):

NotFoundError: /content/dataset/Flickr8k_Dataset/Flicker8k_Dataset1000268201_693b08cb0e.jpg; No such file or directory
[[{{node ReadFile}}]] [Op:IteratorGetNext]