search

mila-iqia / babyai

BabyAI platform. A testbed for training agents to understand and execute language commands.
BSD 3-Clause "New" or "Revised" License
700 stars 146 forks source link

Unable to run PPO agent with arch=bow_endpool_res_pixels #109

Closed ayushchakravarthy closed 3 years ago

ayushchakravarthy commented 3 years ago

Hey! I am new to DeepRL and am implementing a paper in the BabyAI environment. I am trying to run the default ppo algorithm using the train_rl.py script however, I am unable to when I set the flag arch = 'bow_endpool_res_pixels' The error I get is when the image_conv is applied and says that: RuntimeError: Given groups=1, weight of size [128, 3, 8, 8], expected input[64, 128, 56, 56] to have 3 channels, but got 128 channels instead Thanks!

dyth commented 3 years ago

Hello!

The arch name convention is <input>_<config>. So if you want to train from pixels, you would have to use pixels_<config>, We recommend pixels_endpool_res. Hope this helps!

ayushchakravarthy commented 3 years ago

Thank you so much, that worked! Appreciate the help!