Set of tools to run a particular algorithm on a dataset constructed with the validation set constructor, and evaluate it, along with advanced statistics regarding depth value, dans pixel position in image with repsect to flight path vector.
## Inference Example
Get the last frame and a previous frame such that the displacement magnitude is as close to 30cm as possible, with the condition of having a rotation of less that 1 radian. Each frame is preprocessed so that it is of shape `[C, H, W]` and with a range `[0, 1]` instead of `[0, 255]`.
You can find an example usage of this Inference Framework for SfmLearner [here](https://github.com/ClementPinard/SfmLearner-Pytorch/tree/inference-framework)
## Evaluation
The evaluation step is a simple script that takes into input the computed depth maps (here in the file `output.npz`)