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

Jump to bottom

Develop #207

Merged
merged 207 commits into from
Oct 4, 2019
Merged

Develop #207

merged 207 commits into from
Oct 4, 2019

Conversation

ntabris
Copy link
Contributor

@ntabris ntabris commented Oct 4, 2019

No description provided.

ntabris and others added 30 commits September 6, 2019 16:26
@ntabris
include pid in tmp dir name for json.zip
b72c0a2 
this prevents conflicts when multiple processes are using the
same json.zip file, as when training models in parallel.
@ntabris
more tests
2298022
@ntabris
more tests
f80bdee
@ntabris
bug fix to pass new tests
96b1a4a
@ntabris
don't autoconvert to grayscale
7e68b3d
@ntabris
Merge branch 'viz' of https://github.com/murthylab/sleap into viz
1dc7f4a
@ntabris
re-arranged methods in Labels class
3685c80
@ntabris
bug fix when creating Instance w/ zero points
7593282
@ntabris
predict on range of frames
e8151b1
@ntabris
Merge pull request #135 from murthylab/viz
6f8a856 
Viz
@ntabris
opencv-python 3.4.2.17 => 3.4.1.15
41eb461 
This seems to get around issue with missing DLLs.
@ntabris
expose gaussian size and sigma from peak_tf_inference
798235d
@ntabris
bug fixes to last commit
90910e1
@ntabris
determine channels from centroid model
7724818
@ntabris
remove unused crop_iou_threshold
87eaf1d
@ntabris
args to control merge/size for centroid boxes
edb9b42
@ntabris
use / for imgstore paths
9c8ea9f
@ntabris
use / for imgstore paths (bug fix)
0d9b067
@ntabris
Merge pull request #136 from murthylab/viz
0f5f6f6 
Viz
@ntabris
show number of user labeled frames in status bar
e066b39
@ntabris
allow 0-1000 negative samples
dc912b0
@ntabris
show full video filename (don't truncate)
b941894
@ntabris
bug fix, use tolist() to get list of Python ints
e7f37fe
@ntabris
bug fix when confmaps/paf multiscale different
8fe7920
@ntabris
disable blur when upsampling (it's not working)
508b2e4
@ntabris
Merge pull request #146 from murthylab/viz
ff61190 
Viz
@ntabris
support for header (graph), use mark objects
58da178 
also fix track colors so they match instance colors
@ntabris
note change when skeleton added
1d211b0
@ntabris
create output dir if doesn't exist
e503c28
@ntabris
check labels is not None when updating gui
3a091fb
ntabris and others added 29 commits October 1, 2019 10:44
@ntabris
Use "." for paths of hdf5 video in labels file.
fcad548
@ntabris
Release hdf5 labels object in test.
1afe6f0
@ntabris
Close hdf5 videos in test.
5a4a586
@ntabris
Use subset of frames for hdf5 test.
20871d9
@ntabris
Merge pull request #198 from murthylab/viz
998045e 
HDF5 training package
@ntabris
Improve rangelist test coverage.
3462df5
@ntabris
Use player.view in track overlay.
af59969
@ntabris
Make empty frame for empty hdf5 video.
7e20727
@ntabris
Only include negative samples from labeled frames.
688e851
@talmo @ntabris
Update inference overlay to use InferenceModel
cfbee84 
(Merged by ntabris with some additional fixes.)
@ntabris
sLEAP -> SLEAP in readme.
1ffbbd6
@ntabris
LossViewer releases zmq socket when closed.
ddf1d2e
@ntabris
Move LossMonitor test into gui tests dir
cc40440 
so it won't run with gpu tests
@ntabris
Add param to add default suffix when saving.
bcf0470 
When saving labels dataset and we cannot detect valid extension,
the default suffix (if given) will be added.
Training package defaults to saving as h5.
@ntabris
Merge pull request #199 from murthylab/viz
605a2cd 
Viz
@ntabris
Add support for en/decoding frames stored in hdf5.
ac2acba 
Saving labels dataset w/ frames in h5 defaults to png encoding.
@ntabris
Ignore errors when trying to plot initial frame.
d5e4f6e
@ntabris
Merge pull request #201 from murthylab/viz
4cb1a07 
image encoding in HDF5
@ntabris
Add last_frame_idx property on videos.
172fb2b 
For imgstore or hdf5 videos with select frames indexed by the
original video, this lets us show the entire range of frames in the
seekbar for the video.
@ntabris
Fixes so HDF5 selected images -> imgstore works.
20fed20 
to_imgstore now uses (height, width, channels) properties instead
of shape, and HDF5 video casts those properties as ints.
@ntabris
Add missing arg to docstring.
bc899c5
@ntabris
More test coverage for io.video
9e06282
@ntabris
Add __init__.py for sleap.info module.
645c17d
@ntabris
Add __init__.py for sleap.gui.overlays module.
8f021db
@ntabris
Wrappers for QFileDialog functions w/o options arg
bb66140
@ntabris
Merge pull request #203 from murthylab/viz
10923bd 
Fixes for pip install.
@davidt0x
Added support for from_predicted in saved HDF5.
7aded36
@davidt0x
Merge branch 'develop' of https://github.com/murthylab/sleap into bug…
18844c6 
…/from_predicted_hdf5
@davidt0x
Merge pull request #205 from murthylab/bug/from_predicted_hdf5
5a64adc 
Bug/from predicted hdf5
@ntabris ntabris merged commit e44a79f into master Oct 4, 2019
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.

None yet
3 participants
@ntabris @talmo @davidt0x