Easy training on custom dataset. Various backends (MobileNet and SqueezeNet) supported. A YOLO demo to detect raccoon run entirely in brower is accessible at https://git.io/vF7vI (not on Windows).
Thank you for this great work and all the helpful discussions on the issues page. I tried to make predictions by loading the tiny yolo v2 weights trained on coco, by just using the code in this repo. After loading weights, I run the predict (which should work reasonably if the weight loader works correctly?).
I get the following error and no bounding boxes in the predicted image!
C:\Users\bhargav\Documents\phase_1\yolo2-map\utils.py:204: RuntimeWarning: invalid value encountered in subtract x = x - np.max(x) c:\users\bhargav\appdata\local\continuum\miniconda3\envs\tf110gpu\lib\site-packages\numpy\core\_methods.py:29: RuntimeWarning: invalid value encountered in reduce return umr_minimum(a, axis, None, out, keepdims) C:\Users\bhargav\Documents\phase_1\yolo2-map\utils.py:206: RuntimeWarning: invalid value encountered in less if np.min(x) < t: C:\Users\bhargav\Documents\phase_1\yolo2-map\utils.py:88: RuntimeWarning: invalid value encountered in greater netout[..., 5:] *= netout[..., 5:] > obj_threshold C:\Users\bhargav\Documents\phase_1\yolo2-map\utils.py:96: RuntimeWarning: invalid value encountered in greater if np.sum(classes) > 0:
Has anyone tried this? Any help would be appreciated.
bkanaki
commented
6 years ago
Hi,
Thank you for this great work and all the helpful discussions on the issues page. I tried to make predictions by loading the tiny yolo v2 weights trained on coco, by just using the code in this repo. After loading weights, I run the predict (which should work reasonably if the weight loader works correctly?).
I get the following error and no bounding boxes in the predicted image!
C:\Users\bhargav\Documents\phase_1\yolo2-map\utils.py:204: RuntimeWarning: invalid value encountered in subtract x = x - np.max(x) c:\users\bhargav\appdata\local\continuum\miniconda3\envs\tf110gpu\lib\site-packages\numpy\core\_methods.py:29: RuntimeWarning: invalid value encountered in reduce return umr_minimum(a, axis, None, out, keepdims) C:\Users\bhargav\Documents\phase_1\yolo2-map\utils.py:206: RuntimeWarning: invalid value encountered in less if np.min(x) < t: C:\Users\bhargav\Documents\phase_1\yolo2-map\utils.py:88: RuntimeWarning: invalid value encountered in greater netout[..., 5:] *= netout[..., 5:] > obj_threshold C:\Users\bhargav\Documents\phase_1\yolo2-map\utils.py:96: RuntimeWarning: invalid value encountered in greater if np.sum(classes) > 0:Has anyone tried this? Any help would be appreciated.