Skip to content

The interface of DownsampleFactorMax has been changed to Pool #2

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
onehr wants to merge 1 commit into
base: master
Choose a base branch
from
Open

The interface of DownsampleFactorMax has been changed to Pool #2

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
9 changes: 4 additions & 5 deletions src/convnet3d/maxpool3d.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -5,8 +5,7 @@


from theano import tensor
from theano.tensor.signal.downsample import DownsampleFactorMax

from theano.tensor.signal.pool import Pool

def max_pool_3d(input, ds, ignore_border=False):
"""
Expand Down Expand Up @@ -46,7 +45,7 @@ def max_pool_3d(input, ds, ignore_border=False):
input_4D = tensor.reshape(input, new_shape, ndim=4)

# downsample mini-batch of videos in rows and cols
op = DownsampleFactorMax((ds[1],ds[2]), ignore_border)
op = Pool((ds[1],ds[2]), ignore_border)
output = op(input_4D)
# restore to original shape
outshape = tensor.join(0, input.shape[:-2], output.shape[-2:])
Expand All @@ -70,10 +69,10 @@ def max_pool_3d(input, ds, ignore_border=False):
vid_shape), 'int32')
input_4D_time = tensor.reshape(input_time, new_shape, ndim=4)
# downsample mini-batch of videos in time
op = DownsampleFactorMax((1,ds[0]), ignore_border)
op = Pool((1,ds[0]), ignore_border)
outtime = op(input_4D_time)
# output
# restore to original shape (xxx, rows, cols, time)
outshape = tensor.join(0, input_time.shape[:-2], outtime.shape[-2:])
shufl = (list(range(vid_dim-3)) + [vid_dim-1]+[vid_dim-3]+[vid_dim-2])
return tensor.reshape(outtime, outshape, ndim=input.ndim).dimshuffle(shufl)
return tensor.reshape(outtime, outshape, ndim=input.ndim).dimshuffle(shufl)