Metadata-Version: 2.1
Name: ndx-pose
Version: 0.1.1
Summary: NWB extension to store pose estimation data
Home-page: UNKNOWN
Author: Ryan Ly, Ben Dichter, Alexander Mathis
Author-email: rly@lbl.gov, bdichter@lbl.gov, alexander.mathis@epfl.ch
License: BSD 3-Clause
Keywords: NeurodataWithoutBorders,NWB,nwb-extension,ndx-extension
Platform: UNKNOWN
Classifier: Intended Audience :: Developers
Classifier: Intended Audience :: Science/Research
Classifier: License :: OSI Approved :: BSD License
Description-Content-Type: text/markdown; charset=UTF-8
License-File: LICENSE

# ndx-pose Extension for NWB

This is a work in progress and not yet ready for public usage.

## Installation

TBD

## Usage
```python
import datetime
import numpy as np
from pynwb import NWBFile, NWBHDF5IO
from ndx_pose import PoseEstimationSeries, PoseEstimation

nwbfile = NWBFile(
    session_description='session_description',
    identifier='identifier',
    session_start_time=datetime.datetime.now(datetime.timezone.utc)
)

camera1 = nwbfile.create_device(
    name='camera1',
    description='left camera',
    manufacturer='my manufacturer'
)
camera2 = nwbfile.create_device(
    name='camera2',
    description='right camera',
    manufacturer='my manufacturer'
)

data = np.random.rand(100, 3)  # num_frames x (x, y, z)
timestamps = np.linspace(0, 10, num=100)  # a timestamp for every frame
confidence = np.random.rand(100)  # a confidence value for every frame
front_left_paw = PoseEstimationSeries(
    name='front_left_paw',
    description='Marker placed around fingers of front left paw.',
    data=data,
    unit='pixels',
    reference_frame='(0,0,0) corresponds to ...',
    timestamps=timestamps,
    confidence=confidence,
    confidence_definition='Softmax output of the deep neural network.',
)

data = np.random.rand(100, 2)  # num_frames x (x, y)
timestamps = np.linspace(0, 10, num=100)  # a timestamp for every frame
confidence = np.random.rand(100)  # a confidence value for every frame
front_right_paw = PoseEstimationSeries(
    name='front_right_paw',
    description='Marker placed around fingers of front right paw.',
    data=data,
    unit='pixels',
    reference_frame='(0,0,0) corresponds to ...',
    timestamps=front_left_paw,  # link to timestamps of front_left_paw
    confidence=confidence,
    confidence_definition='Softmax output of the deep neural network.',
)

pose_estimation_series = [front_left_paw, front_right_paw]

pe = PoseEstimation(
    pose_estimation_series=pose_estimation_series,
    description='Estimated positions of front paws using DeepLabCut.',
    original_videos=['camera1.mp4', 'camera2.mp4'],
    labeled_videos=['camera1_labeled.mp4', 'camera2_labeled.mp4'],
    dimensions=np.array([[640, 480], [1024, 768]], dtype='uint8'),
    scorer='DLC_resnet50_openfieldOct30shuffle1_1600',
    source_software='DeepLabCut',
    source_software_version='2.2b8',
    nodes=['front_left_paw', 'front_right_paw'],
    edges=np.array([[0, 1]], dtype='uint8'),
    # devices=[camera1, camera2],  # this is not yet supported
)

behavior_pm = nwbfile.create_processing_module(
    name='behavior',
    description='processed behavioral data'
)
behavior_pm.add(pe)

path = 'test_pose.nwb'
with NWBHDF5IO(path, mode='w') as io:
    io.write(nwbfile)

with NWBHDF5IO(path, mode='r', load_namespaces=True) as io:
    read_nwbfile = io.read()
    read_pe = read_nwbfile.processing['behavior']['PoseEstimation']
    print(read_pe)
```


## Contributors
- @rly
- @bendichter
- @AlexEMG

This extension was created using [ndx-template](https://github.com/nwb-extensions/ndx-template).


