Skip to content

Commit

Permalink
Need to optimize memory, add a way to have multiple runs
Browse files Browse the repository at this point in the history
  • Loading branch information
IshanManchanda committed May 23, 2020
1 parent f68763a commit 45ffe3c
Show file tree
Hide file tree
Showing 2 changed files with 2 additions and 1 deletion.
2 changes: 1 addition & 1 deletion digits.py
Original file line number Diff line number Diff line change
Expand Up @@ -14,7 +14,7 @@ def main():
deskew_data()

training, validation, test = load_data()
train([784, 128, 10], 0.008, 0, 0.05, training, validation)
train([784, 128, 10], 0.008, 0.2, 0.05, training, validation)


# root = tk.Tk()
Expand Down
1 change: 1 addition & 0 deletions src/utils.py
Original file line number Diff line number Diff line change
Expand Up @@ -15,6 +15,7 @@ def load_data(deskew=True):
Returns the training, validation, and test data as a list of tuples,
where the outputs are one-hot encoded vectors.
"""
# TODO: Optimize memory usage by loading only parts.

# The _py3 version of the dataset is a redumped version for Python 3
# which doesn't use Python 2's latin1 encoding
Expand Down

0 comments on commit 45ffe3c

Please sign in to comment.