Skip to content
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

Adds constant memory backprop #2

Open
wants to merge 15 commits into
base: master
Choose a base branch
from
Open

Adds constant memory backprop #2

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
15 commits
Select commit Hold shift + click to select a range
55b19a8
Adds gradient computations for layers
anshuln Jul 13, 2019
1a8d7c8
Adds const memory backprop computations for layers
anshuln Jul 13, 2019
d3ee0c2
Merge upstream master
anshuln Jul 13, 2019
fb1bb54
Merge
anshuln Jul 13, 2019
97bf377
Adds fit function, needs some testing
anshuln Jul 13, 2019
d096c5d
Merge commit
anshuln Jul 15, 2019
85fe171
Adds const memory backprop for AffineCoupling and other layers. Howev…
anshuln Jul 15, 2019
1c95d38
Adds custom fit funtion using ndarray or tf Tensor
anshuln Jul 15, 2019
d897faf
Adds unit tests for gradient descent
anshuln Jul 16, 2019
eefd903
Adds unit tests for gradient descent
anshuln Jul 16, 2019
990b302
Fixes incorrect gradient for AffineCoupling
anshuln Jul 16, 2019
35a135e
Adds fit_generator for models
anshuln Jul 17, 2019
70dfb23
Minor bug fixes
anshuln Jul 18, 2019
74904a0
Potential fix for incorrect dy issue
anshuln Jul 18, 2019
62b8822
Fixes order of gradients in tests
anshuln Jul 18, 2019
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
20 changes: 20 additions & 0 deletions invtf/data.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,20 @@
import tensorflow as tf
import os

def load_image_dataset(folder, epochs=1,new_image_size=(64, 64),batch_size=32,shuffle=True,):
def _parse_function(filename):
image_string = tf.io.read_file(filename)
image_decoded = tf.image.decode_jpeg(image_string)
image_resized = tf.image.resize(image_decoded, new_size)
return image_resized

files = ['{}/{}'.format(folder,f) for f in os.listdir(
folder) if os.path.isfile(os.path.join(folder, f))]
dataset = tf.data.Dataset.from_tensor_slices(tf.constant(files))
if shuffle == True:
dataset = dataset.shuffle(buffer_size=100)
dataset = dataset.repeat(count=epochs)
dataset = dataset.map(map_func=_parse_function,num_parallel_calls=4)
dataset = dataset.prefetch(buffer_size=batch_size)
dataset = dataset.batch(batch_size=batch_size)
return iter(dataset),(len(files)//batch_size)
Loading