SamThomas
SamThomas
I would think that the gradient calculation is identical to Voxblox. See [interpolator_inl.h](https://github.com/ethz-asl/voxblox/blob/7f7892de15c7760ce0fd777315fd1f74f73eda02/voxblox/include/voxblox/interpolator/interpolator_inl.h#L47)
Have a look at the end of the main function in inference.py Change plt.axis('off') plt.imshow(mask) plt.savefig(path_to_output, bbox_inches='tight') To mask.save(path_to_output) As for training with your own dataset, Search the entire project...
I have modified the values of PASCAL_VOC_classes in utils.py with 16 new values for my own dataset. When I try to train I get the error: `tensorflow.python.framework.errors_impl.InvalidArgumentError: Dimension 0 in...
Am also seeing this.
Reducing my learning rate solved it.
I found my error. I had some training data with values way outside me expected range of 0-1 which I found by adding an assert in my dataloader.
Did you run it from the command line like this? `python youtube.py 300 urls.txt 38 65`
If you only want to be able to display WGS84 coordinates, the simplest way I can think of would be to provide a view origin (lat, lon, alt) that becomes...
If you want to display map tiles from a WMS, you would need to implement [slippy map](https://en.wikipedia.org/wiki/Tiled_web_map) rendering.