Skip to content

Update mknet.py #8

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

Update mknet.py #8

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
24 changes: 12 additions & 12 deletions lsd/tutorial/example_nets/fib25/acrlsd/mknet.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,7 +1,7 @@
import json
import mala
from funlib.learn.tensorflow import models
from funlib.learn.tensorflow.models.unet import crop
import tensorflow as tf
from mala.networks.unet import crop_zyx

def create_auto(input_shape, output_shape, name):

Expand All @@ -12,20 +12,20 @@ def create_auto(input_shape, output_shape, name):
raw = tf.placeholder(tf.float32, shape=input_shape)
raw_batched = tf.reshape(raw, (1, 1) + input_shape)

unet, _, _ = mala.networks.unet(
unet, _, _ = models.unet(
raw_batched,
12,
6,
[[2,2,2],[2,2,2],[3,3,3]])

embedding_batched, _ = mala.networks.conv_pass(
embedding_batched, _ = models.conv_pass(
unet,
kernel_sizes=[1],
num_fmaps=10,
activation='sigmoid',
name='embedding')

embedding_batched = crop_zyx(embedding_batched, (1, 10) + output_shape)
embedding_batched = crop(embedding_batched, (1, 10) + output_shape)
embedding = tf.reshape(embedding_batched, (10,) + output_shape)

print("input shape : %s"%(input_shape,))
Expand All @@ -50,21 +50,21 @@ def create_affs(input_shape, intermediate_shape, expected_output_shape, name):
raw = tf.placeholder(tf.float32, shape=input_shape)
raw_batched = tf.reshape(raw, (1, 1) + input_shape)
raw_in = tf.reshape(raw_batched, input_shape)
raw_batched = crop_zyx(raw_batched, (1, 1) + intermediate_shape)
raw_batched = crop(raw_batched, (1, 1) + intermediate_shape)
raw_cropped = tf.reshape(raw_batched, intermediate_shape)

pretrained_lsd = tf.placeholder(tf.float32, shape=(10,) + intermediate_shape)
pretrained_lsd_batched = tf.reshape(pretrained_lsd, (1, 10) + intermediate_shape)

concat_input = tf.concat([raw_batched, pretrained_lsd_batched], axis=1)

unet, _, _ = mala.networks.unet(
unet, _, _ = models.unet(
concat_input,
12,
6,
[[2,2,2],[2,2,2],[3,3,3]])

affs_batched, _ = mala.networks.conv_pass(
affs_batched, _ = models.conv_pass(
unet,
kernel_sizes=[1],
num_fmaps=3,
Expand Down Expand Up @@ -129,15 +129,15 @@ def create_config(input_shape, output_shape, name):

if __name__ == "__main__":

train_input_shape = (304, 304, 304)
train_input_shape = (328, 328, 328)
train_intermediate_shape = (196, 196, 196)
train_output_shape = (92, 92, 92)
train_output_shape = (72, 72, 72)

create_auto(train_input_shape, train_intermediate_shape, 'train_auto_net')
create_affs(train_input_shape, train_intermediate_shape, train_output_shape, 'train_net')

test_input_shape = (364, 364, 364)
test_output_shape = (260, 260, 260)
test_output_shape = (240, 240, 240)

create_affs(test_input_shape, test_input_shape, test_output_shape, 'test_net')

Expand Down