-
Intro:
- When dealing with large datasamples, loading data from imagefile/excel file in to numpy makes training slow.
- This tutorial creates a hdf5 database from the samples. The hdf5 contains train , test datasets.
- To know about how to use the hdf5 file for the training data loader, check the repo https://github.com/sujayr91/TimeSeries_Classification_LSTM
-
Requirements:
- Install h5py package: pip install h5py
- python 3
-
Usage:
- Organize data in to folders for each label[this assumes data samples are in csv, for image data change file load ex: PIL.Image]
- Use createhdf5database.py for creating hdf5 datasets.
-
Notifications
You must be signed in to change notification settings - Fork 0
sujayr91/HDF5-database-for-deeplearning
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Folders and files
Name | Name | Last commit message | Last commit date | |
---|---|---|---|---|
Repository files navigation
About
Building train/test datasets with HDF5 database for deeplearning
Topics
Resources
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published