search

kentsommer / pytorch-value-iteration-networks

Pytorch implementation of Value Iteration Networks (NIPS 2016 best paper)
BSD 3-Clause "New" or "Revised" License
318 stars 62 forks source link

RUN ERROR #7

Closed N-Kingsley closed 4 years ago

N-Kingsley commented 6 years ago

when I run 'python train.py --datafile dataset/gridworld_8x8.npz --imsize 8 --lr 0.005 --epochs 30 --k 10 --batch_size 128', it's ok,but again 'python train.py --datafile dataset/gridworld_16x16.npz --imsize 16 --lr 0.002 --epochs 30 --k 20 --batch_size 128' was run, an error occurred as follows: ni@ubuntu:~/pytorch-value-iteration-networks$ python train.py --datafile dataset/gridworld_16x16.npz --imsize 16 --lr 0.002 --epochs 10 --k 20 --batch_size 128 Traceback (most recent call last): File "train.py", line 135, in config.datafile, imsize=config.imsize, train=True, transform=transform) File "/home/ni/pytorch-value-iteration-networks/dataset/dataset.py", line 22, in init self._process(file, self.train) File "/home/ni/pytorch-value-iteration-networks/dataset/dataset.py", line 58, in _process images = images.astype(np.float32) MemoryError

michaalbert commented 6 years ago

I get the same traceback when trying to train on the 28x28 domain. If the computer I am running it on didn't have 32GB RAM I'd blame it on not having enough RAM, but that seems very unlikely. Any idea where to even start debugging would be much appreciated.

chachabules commented 5 years ago

the memory error problem means that your computer don't have enough memory for this program because numpy need much memory to store the arrays. It is to add some memory for virtual memory for your linux system or windows. Your can also reduce the sample number. Hope to help you.

kentsommer commented 4 years ago

Closing out as there are no new comments for a while and I agree that this is likely an issue of simply running out of memory.