leonl42
Results
2
issues of
leonl42
Hey, In your paper you report that "agents achieve an accuracy of 87.36%, 93.25%, 96.09% when trained on 64, 256, and 1024 candidates respectively". However, if I analyzed figure 2...
In line 198 in train_kmeans.py: `kmeans_dict = {'rot': kmeans_rot_q, 'scale': kmeans_sc_q, 'sh': kmeans_sh_q, 'dc': kmeans_dc_q}` The problem with this line is that it accesses variables that are never defined if...