Skip to content

Conversation

@Mostafa3zazi
Copy link
Member

No description provided.

Then you can use my script above for training.

## I want to fine-tune CNN part.
Officially, this doesn’t support the CNN tuning. That’s what I did in the original paeper. However, informally, I made it to train CNN part too... but I didn't document it, now, after two years, I don't remember very much. `train_image_caption_model.py` is the script to train the CNN part. I also remember that I tried to use another preprocessed json than I document here. Currently I have the `_dic` file and the main processed file separately but I combined them. The script to generate a preprocessed file in the new format (i.e that one compatible with `train_image_caption_model.py` ) should be `code/preprocess_captions.py`. That's what I vaguely remember.
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Add your own README file, and reference the original file with proper credits.

"fenced-in": 8513,
"fences": 8720,
"fencing": 7675,
"fende
Copy link
Member

@mhashim6 mhashim6 Feb 24, 2020

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Provide an Arabic version of this file, you can use this code for translation.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants