Rodrigo Benenson
Rodrigo Benenson
I think `propose_hf5.py` has what you are looking for. One the input options is `-i A List of images to evaluate`, from what I can see the images path is...
Using `python propose_hf5.py ../models/lpo_VOC_0.05.dat ~/Desktop/output.hf5 -b -i ~/Desktop/im1.png` solved the "too few arguments" error for me on python3.4
why not have humane names, and have `__init__.py` import the relevant files. so `mnist/__init__.py` would become `mnist/mnist.py` and `mnist/__init__.py` would be a single liner containing `from .mnist import *`. Would...
https://github.com/jgarff/rpi_ws281x works fine for me on the RPi 2.
Another option would be to have an argument in tfds.as_numpy() that indicates how to treat RaggedTensors (with default to "return as-is"). Without it I resorted to implementing my own versions...
Indeed that seems to fit better my use case. Then I suggest that tfds.as_numpy(...) docstring should mention something along the lines of "for eager mode only use cases, ds.as_numpy_iterator(...) might...
The tau selection/training is described here https://openreview.net/forum?id=r1gRTCVFvB¬eId=SJx9gIcsoS no clue about the code.
Any clue of what could the cause of this issue ? How should I approach it ? Any suggested work around (other than not using polymer) ?
What do you mean by "active transition" ? Even when waiting long time between step 2 and 3 the issue appears.