Firstly you will find an attached document called paper overview which is a simple overview few about paper "multimodal few-shot learning with frozen language models" Paper Link https://papers.nips.cc/paper/2021/file/01b7575c38dac42f3cfb7d500438b875-Paper.pdf
The Fashion-MNIST clothing classification problem is a new standard dataset used in computer vision and deep learning.
Dataset Link https://github.com/zalandoresearch/fashion-mnist
Although the dataset is relatively simple, it can be used as the basis for learning and practicing how to develop, evaluate, and use deep convolutional neural networks for image classification from scratch, Which achieves impressive results.
1- Download Dataset
2- Using Convoultion Neural Network for training the data
3- Calculate Receptive field
4- Calculate FLOPS and MAACs
Receptive field
1- https://github.com/shelfwise/receptivefield
2- https://github.com/google-research/receptive_field
FLOPS and MAACs
1- https://github.com/ckyrkou/Keras_FLOP_Estimator
1- pip install -r requirements.txt
2- python main.py