DCGAN #389
Open
DCGAN #389
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…any mode. When using mnist dataset and c_dim equals 1, the shape of value of merge(images, size) is (224, 224, 1), while the function scipy.misc.imsave just support these three type of shape: MxN or MxNx3 or MxNx4(https://docs.scipy.org/doc/scipy/reference/generated/scipy.misc.imsave.html). So we need to squeeze it first.
Revert "Use visualize option."
Deel with ValueError in utils.imsave when using mnist.
I have attempted to use the program with grayscale images. c_dim will become n for a n x n image (e.g. c_dim = 96 for a 96 x 96 image). By modifying as such, grayscale image could be successfully detected and c_dim will be appropriately assigned with 1.
Bug Fix for Checking Grayscale Image
Fix layer 4 name in discriminator
Remove unused placeholder variable in build_model
fix generation of sample image non square number of images
Fixed manifold dimension bug
Update utils.py
combined code paths for with/without y_dim
Rename `15:07:47` to `15_07_47` in test_***.png files so that file names would be legal on Windows
Make dataset root directory a flag
Shuffle self.data after assignment
prevent xrange error: float cannot be interpreted as an integer
Give informative error msg when no data exists
Raise error if dataset size less than batch_size
Update model.py
Add informative msg when img size mismatch
… normal; 'uniform_signed': (-1,1); 'uniform_unsigned':(0,1) }
…ncy with --ckpt_freq
- set root output directory with --out_dir. - set output name with --out_name (or leave blank for it to be generated automatically from hyperparams) - checkpoints, samples and log saved under out_dir/out_name - environment variables expanded - save json for FLAGS for reference - auto set output size to input size if None
move self.data length check to the right place
…e 1-channel images
Fix 3-channel images incorrectly read as if 1-channel
A number of extra options
Silly mistake!
Changed resize function; scipy deprecated Issue #351
Add a missing prerequisite in Readme.md
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
edits