search

suraj-deshmukh / Active-Learning

Active Learning using Multi Label Image Dataset
https://suraj-deshmukh.github.io/Active-Learning/
14 stars 4 forks source link

Convert my own data into .mat file #1

Open rakashi opened 6 years ago

rakashi commented 6 years ago

Hi I am trying to convert my own dataset into .h5 and file but I am not having .mat file how can I convert into .mat file and .h5 file

suraj-deshmukh commented 6 years ago

The dataset link is given in Dataset section. Download the dataset and extract it and follow the mentioned procedure given in Dataset section of this repository.

rakashi commented 6 years ago

thanks for your reply I have downloaded that dataset but that contains only natural scenes(about 2000 images) but I am having 1000 classes of images. I want to convert this dataset into .mat, .h5 file. How can I convert this 1000 classes of images I have seen generate-data.py script but it is asking .mat file If I want to convert my own dataset

suraj-deshmukh commented 6 years ago

You don't need to convert your dataset into .mat format. You can easily convert your dataset into .h5 format. For eg. F = h5py.File("dataset.h5") F['target'] = "your target lables in numpy array" F.close()

suraj-deshmukh commented 6 years ago

You can't use this repository as it is for your project. You need to do some changes according to your project.

rakashi commented 6 years ago

thanks for your reply I am trying to find out multilabel classification using 100 or 1000 classes of image Here I am trying to use as your code for converting .h5 file by keeping target lables as same your .mat file Is it right way ?