Image Captioning

#1

My model is not able to predict the sentence for given image. it’s just only predicting the endseq everytime.
Here’s the code link:-
https://drive.google.com/file/d/1x5SSjAqGtSCPDvP2LQyp6OfWxf5aHE1w/view?usp=sharing

#2

Hey @parikshitagarwal40,
I did a few changes to your code, Can you please run and check for the output. Since I don’t have the weights with me I can’t run it in real-time.

#3

still not predicting anything,

#4

I’m sharing my model weights and word_idx_dict you please load these and run it in real-time:- , word_idx_dictonary:- https://drive.google.com/file/d/1RMovyKvNpOUWYpF_hdYdV7wBkwOdF8lV/view?usp=sharing, model_weights:- https://drive.google.com/file/d/1ljhmR0D1G0EPbDEnmo_2Jy7DPRhsgkdC/view?usp=sharing

#5

file

This is a working file try debugging your code with this while I’m trying out your code.

#6

I tried to run through your code using the flickr_8k dataset which I had and It works fine there is some problem with your model is not trained well.
Try to re-train it.