Classifying 8 different dance forms using transfer learning, while utilizing pre-trained model ResNet50.
This International Dance Day, an event management company organized an evening of Indian classical dance
performances to celebrate the rich, eloquent, and elegant art of dance.
The task is to build a deep learning model that can help the company classify these images into
eight categories of Indian classical dance.
The eight categories of Indian classical dance are as follows:
This data set consists of the following two columns:
Column Name | Description |
---|---|
Image | Name of Image |
Target | Category of Image ['manipuri','bharatanatyam','odissi','kathakali','kathak','sattriya','kuchipudi','mohiniyattam'] |
The data folder consists of two folders and two .csv files. The details are as follows:
Given below shows the basic architecture implemented along with ResNet50.
The test accuracy finally received is 82.5%.
As the number of training images are only 356 and the number of classes are 8, 82.5% seems to be
a decent accuracy. Image augmentation played a major role in the entire training.
Exact code can be seen in danceForms.ipynb file.