Tessa van der Heiden

icon indicating an email [email protected]

icon company NPO icon location Hilversum icon location Currently working at NPO, the public broadcasting service of the Netherlands, as a Machine Learning Engineer.

Results 5 comments of Tessa van der Heiden

Segmentation fault (core dumped)

I run it on Ubuntu and dont have this problem.

Checking collisions with agent itself?

Hi! No, because it will just add a constant negative reward everytime, so the gradient is not affected ;-)

pip does not support tensorflow installation requirements

Hi Felix, So, it would be great to work with newer versions of all packages and Python (just to make it compatible with more user's environments). I've changed the "tf."...

BENCH: optimize: add timings to global optimizers benchmarks

Hey @lucascolley, Thanks for your comment! I've posted a [message](https://mail.python.org/archives/list/[email protected]/thread/3VW5O7AYDIGAOMPGVEGWJ35GJZ27XLR5/), but I am not sure if it resulted in much traction. Any ideas how we can merge it in?

Dynamically changing maze worlds

Yes, something like this below? But where to put it, in the `act()`? def remove_wall(self, cell, direction): self.adjacencies[cell[0]][cell[1]].append(direction) # remove opposite action new_cell = cell + self.actions[direction] self.adjacencies[new_cell[0]][new_cell[1]].append(self.opposite[direction]) # remove...