Skip to main content

No project description provided

Project description

AER-toolbox

This library intends to be a minimal tool for loading events from files with common event-camera file extensions into Python.

See the project on PyPI or do pip3 install aertb

Usage

from aertb.core import FileLoader

datLoader = FileLoader('dat') # 'bin', or 'aedat'
datLoader.load_events('../example_data/dat/cars/obj_004414_td.dat')

Supported extensions:

  • .dat: N-Cars / Prophesee Cameras
  • .bin: N-MNIST, N-Caltech101
  • .aedat: PokerDVS
  • .mat: DVS-Barrel

It also make the process of loading and iterating HDF5 files easier.

from aertb.core import HDF5File

dataset_train = HDF5File('TRAIN.h5')
train_iterator = dataset_train.iterator(n_samples_group=10, rand=23)

for sample in tqdm(train_iterator):
    # do something with sample.events, sample.label or sample.name

Example: making a GIF

from aertb.core import HDF5File, make_gif

file = HDF5File('../DVS_Barrel.hdf5')
sample = file.load_events(group='moving', name='11')
make_gif(sample, filename='sample_moving.gif', camera_size=(128, 128), n_frames=480, gtype='std')

The library also includes a command line interface for converting files from a given extension to hdf5, as well as gif making capabilities for easy visualisation of the files.

Opening the CLI

  1. If the install with pip worked perfectly, you can now type aertb in a terminal window and the CLI will open.

  2. If you are installing it from Github: download you should download the project from github and follow the following instructions:

    • a) git clone ...
    • b) Create a virual environment, if venv is not installed run pip install virtualenv, then python3 -m venv aertb_env
    • c) Run source aertb_env/bin/activate
    • d) Run the following command: pip install -r requirements.txt
    • e) Open the cli with python3 . or with the __main__.py file

Using the CLI

  1. Once the CLI is open you get a a similar output on your terminal: Cli Animation
  2. type help to see supported commands and help <topic> to get more info of the command

Examples:

Creating an HDF5 out of a directory

tohdf5 -f 'example_data/dat' -e 'dat' -o 'mytest.h5'

The recommended directory shape is :

 |--Parent (given as parameter)
      |-- LabelClass1
          |-- SampleName1
          |-- SampleName2
          |-- ....
      |-- LabelClass2
          |-- SampleName1
          |-- SampleName2
          |-- ....
      |-- ...

And we suggest that train and test are kept as separate folders so they translate to two different files

Creating an HDF5 out of a single file

tohdf5 -f 'example_data/bin/one/03263.bin' -o 'mytest2.h5'

Creating a gif out of a given file

makegif -f 'example_data/prophesee_dat/test_23l_td.dat' -o 'myGif.gif' -nfr 240 -g 'std'

Gif Animation

Exiting the CLI:

  1. type quit
  2. Exit virtual environment: $ deactivate

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

aertb-0.4.1.tar.gz (26.5 kB view details)

Uploaded Source

File details

Details for the file aertb-0.4.1.tar.gz.

File metadata

  • Download URL: aertb-0.4.1.tar.gz
  • Upload date:
  • Size: 26.5 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.2.0 pkginfo/1.5.0.1 requests/2.24.0 setuptools/49.2.0.post20200714 requests-toolbelt/0.9.1 tqdm/4.47.0 CPython/3.8.3

File hashes

Hashes for aertb-0.4.1.tar.gz
Algorithm Hash digest
SHA256 be3205e93dce81de7e534e0d9b1991de75e9eccbb9bb324b600b63a3aca99de8
MD5 5ecb3105675b6d03121263dd951db8e1
BLAKE2b-256 d499323d032de4e9c2b6db3e9b05bcf74c5cc221c070ceb08c606561422c7980

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page