LAB2 - Hiresh12/UMKC GitHub Wiki

BIG DATA ANALYTICS AND APPLICATIONS

Hiresh Jakkala Bhaskar – 8 Anvesh Mandadi – 17

AIM :

To Generate 4 captions for each image in our dataset using Show and Tell Model and report the precision of the model in BLEU, CIDER, METEOR and ROGUE measures.

Code Implementation :

Importing show and tell model

Iterating each image in caption text file and generating caption for the image using Show and Tell model and storing the actual and expected captions in a variable

Output:

Sending the expected and actual captions of the images to the BLEU, CIDER, METEOR and ROGUE for calculating the precision of the model.

Output:

Steps to execute:

• Copy captions to the caption file

• Copy respective images in imgs folder

• Run the Homefile.py

Code for Accuracy measure:

Meteor

BLEU

CIDER

ROGUE

Video

https://youtu.be/8hq7Z4AGpmk

Reference

https://github.com/salaniz/pycocoevalcap