matchstic / depthmapper Goto Github PK

I am using jetson nano to compute stereo disparity maps using your script. But it fails to compute calibration. Please let me know how to fix this.

Thanks in advance.

Hit any key when you have the chessboard prepared
Calibration :: Computing...
Traceback (most recent call last):
File "calibrate.py", line 20, in
main()
File "calibrate.py", line 17, in main
calibrator.compute_calibration(True)
File "/home/madara/Documents/depthmapper/lib/calibration.py", line 68, in compute_calibration
calibrator = StereoCalibrator( self.chessboard_rows, self.chessboard_cols, self.chessboard_size, (self.width, self.height))
File "/home/madara/.local/lib/python3.6/site-packages/stereovision/calibration.py", line 190, in init
corner_coordinates = np.zeros((np.prod(pattern_size), 3), np.float32)
File "<array_function internals>", line 6, in prod
File "/home/madara/.local/lib/python3.6/site-packages/numpy/core/fromnumeric.py", line 3000, in prod
keepdims=keepdims, initial=initial, where=where)
File "/home/madara/.local/lib/python3.6/site-packages/numpy/core/fromnumeric.py", line 87, in _wrapreduction
return ufunc.reduce(obj, axis, dtype, out, **passkwargs)
TypeError: cannot perform reduce with flexible type

when i run this command :

conda env create -f environment.yml

i got an error like this :

Collecting package metadata (repodata.json): done
Solving environment: failed

ResolvePackageNotFound: 
  - pytorch=1.2.0
  - torchvision=0.4.0
  - cudatoolkit=10.0
  - cudnn=7.6.5

I don't know what the issue is so I come to find some help