bob issueshttps://gitlab.idiap.ch/groups/bob/-/issues2018-02-22T10:37:55Zhttps://gitlab.idiap.ch/bob/docs/-/issues/1Some figures of bob.bio.base do not show up2018-02-22T10:37:55ZAmir MOHAMMADISome figures of bob.bio.base do not show upsee https://www.idiap.ch/software/bob/docs/bob/docs/master/bob.bio.base/doc/struct_bio_rec_sys.htmlsee https://www.idiap.ch/software/bob/docs/bob/docs/master/bob.bio.base/doc/struct_bio_rec_sys.htmlhttps://gitlab.idiap.ch/bob/bob.bio.vein/-/issues/16Not ported to the new CI2018-02-22T11:21:29ZTiago de Freitas PereiraNot ported to the new CIConda-based CIAndré AnjosAndré Anjoshttps://gitlab.idiap.ch/bob/bob.pad.base/-/issues/14guide for creating file-list based databases2018-02-22T11:31:34ZAmir MOHAMMADIguide for creating file-list based databasesIt used to be here:
https://www.idiap.ch/software/bob/docs/bob/bob.pad.db/master/guide.htmlIt used to be here:
https://www.idiap.ch/software/bob/docs/bob/bob.pad.db/master/guide.htmlhttps://gitlab.idiap.ch/bob/bob.pad.base/-/issues/12Nightlies2018-02-22T11:31:34ZTiago de Freitas PereiraNightliesOur nightlies are broken,
https://gitlab.idiap.ch/bob/bob.nightlies/-/jobs/99181
https://gitlab.idiap.ch/bob/bob.nightlies/-/jobs/99185
Could you please have a look @amohammadi since this is your contribution?Our nightlies are broken,
https://gitlab.idiap.ch/bob/bob.nightlies/-/jobs/99181
https://gitlab.idiap.ch/bob/bob.nightlies/-/jobs/99185
Could you please have a look @amohammadi since this is your contribution?https://gitlab.idiap.ch/bob/bob.pad.base/-/issues/9requirements.txt and develop.cfg are incomplete2018-02-22T12:48:36ZManuel Günthersiebenkopf@googlemail.comrequirements.txt and develop.cfg are incompleteIn the `develop.cfg`, only packages with a direct dependency are listed. This is contrary to our default policy, where all dependent packages are listed, recursively and in the correct order.In the `develop.cfg`, only packages with a direct dependency are listed. This is contrary to our default policy, where all dependent packages are listed, recursively and in the correct order.Manuel Günthersiebenkopf@googlemail.comManuel Günthersiebenkopf@googlemail.comhttps://gitlab.idiap.ch/bob/bob.bio.base/-/issues/104mention bob.bio.vein in the documentation too2018-02-25T19:58:48ZAmir MOHAMMADImention bob.bio.vein in the documentation tooping @andre.anjosping @andre.anjosConda-based CIAndré AnjosAndré Anjoshttps://gitlab.idiap.ch/bob/nightlies/-/issues/28Mention which packages are private2018-02-26T07:53:13ZAmir MOHAMMADIMention which packages are privateWhen we do a release, I usually go through the list of packages here and release them.
It would be nice to know which ones are private so that I don't release them.When we do a release, I usually go through the list of packages here and release them.
It would be nice to know which ones are private so that I don't release them.https://gitlab.idiap.ch/bob/nightlies/-/issues/19Some librarires compile twice2018-02-26T07:54:05ZAmir MOHAMMADISome librarires compile twiceSome libraries like `bob.io.base` and `bob.math` compile twice during these two commands:
```sh
$ ./bin/buildout
$ ./bin/python setup.py bdist_wheel
```
You can see this in the logs here: https://gitlab.idiap.ch/bob/bob.nightlies/builds/...Some libraries like `bob.io.base` and `bob.math` compile twice during these two commands:
```sh
$ ./bin/buildout
$ ./bin/python setup.py bdist_wheel
```
You can see this in the logs here: https://gitlab.idiap.ch/bob/bob.nightlies/builds/39385
I guess this is not a problem with nightlies by itself but I am reporting it here.
ping @andre.anjoshttps://gitlab.idiap.ch/bob/bob.pad.face/-/issues/5remove the version script2018-02-28T15:30:41ZAmir MOHAMMADIremove the version scriptit's coming from the templateit's coming from the templateOlegs NIKISINSOlegs NIKISINShttps://gitlab.idiap.ch/bob/nightlies/-/issues/33bob.io.image broken at nightlies2018-02-28T17:47:06ZAndré Anjosbob.io.image broken at nightliesThe current version of bob.io.image is not compiling properly at the nightlies and that is despite the fact nothing really changed that could break it.
The problem happens on all architectures and Python variants (see, e.g.: https://git...The current version of bob.io.image is not compiling properly at the nightlies and that is despite the fact nothing really changed that could break it.
The problem happens on all architectures and Python variants (see, e.g.: https://gitlab.idiap.ch/bob/bob.nightlies/-/jobs/125060)
The error:
```text
OSError: libgif.so.7: cannot open shared object file: No such file or directory
```
What is strange:
* bob.io.image has not changed since last week where we managed to build it successfuly
* There are no changes on its dependencies and giflib, in particular giflib 1.5.4-3 has not changed for a few months
Any clues?Conda-based CIAndré AnjosAndré Anjoshttps://gitlab.idiap.ch/bob/bob.pad.base/-/issues/17Add clients option to driver of FileList db2018-03-01T11:14:25ZPavel KORSHUNOVAdd clients option to driver of FileList dbIt can be useful to print out the list of clients in a given filelist databaseIt can be useful to print out the list of clients in a given filelist databasehttps://gitlab.idiap.ch/bob/bob.pad.face/-/issues/11Converting frame containers to numpy arrays are useful for everyone.2018-03-02T08:28:42ZAmir MOHAMMADIConverting frame containers to numpy arrays are useful for everyone.I can see that why a function which takes a frame container and converts it to a numpy array could be useful here but it is also useful in all packages that use `bob.bio.video`. Hence, this is better to be implemented in `bob.bio.video`....I can see that why a function which takes a frame container and converts it to a numpy array could be useful here but it is also useful in all packages that use `bob.bio.video`. Hence, this is better to be implemented in `bob.bio.video`.
https://gitlab.idiap.ch/bob/bob.pad.face/blob/b0a14393f109e8bc15928ade60b0614e34e4b73f/bob/pad/face/algorithm/VideoSvmPadAlgorithm.py#L116https://gitlab.idiap.ch/bob/bob.db.uvad/-/issues/1Not ported to the new CI2018-03-03T08:39:51ZTiago de Freitas PereiraNot ported to the new CIConda-based CIAmir MOHAMMADIAmir MOHAMMADIhttps://gitlab.idiap.ch/bob/bob.db.oulunpu/-/issues/2Not ported to the new CI2018-03-03T10:40:04ZTiago de Freitas PereiraNot ported to the new CIConda-based CIAmir MOHAMMADIAmir MOHAMMADIhttps://gitlab.idiap.ch/bob/bob.db.batl/-/issues/1package depends on batl.utils2018-03-07T12:47:49ZOlegs NIKISINSpackage depends on batl.utils@salim.kayal. I noticed that this package depends on batl.utils , which is not reflected in requirements.txt . One solution is just to update the requirements. **But**, do we want this package to be dependent on batl.utils? May-be it is ...@salim.kayal. I noticed that this package depends on batl.utils , which is not reflected in requirements.txt . One solution is just to update the requirements. **But**, do we want this package to be dependent on batl.utils? May-be it is better to avoid this dependency and relocate a couple of methods from batl.utils to bob.db.batl ? @salim.kayal, what do you think?https://gitlab.idiap.ch/bob/bob.pad.face/-/issues/20Remove VideoDataLoader.py class2018-03-09T11:25:58ZOlegs NIKISINSRemove VideoDataLoader.py classThis class can now be removed, because it was used only in VideoQualityMeasure and VideoLBPHistogram, which are deleted.This class can now be removed, because it was used only in VideoQualityMeasure and VideoLBPHistogram, which are deleted.Olegs NIKISINSOlegs NIKISINShttps://gitlab.idiap.ch/bob/bob.pad.face/-/issues/18Possible duplication of database configuration files.2018-03-09T11:25:59ZOlegs NIKISINSPossible duplication of database configuration files.As @heusch pointed out, we have similar database configuration files in two places. Initially I suggested just to delete identical files, but after a closer look I realised that files are actually different.
For example we have:
bob/p...As @heusch pointed out, we have similar database configuration files in two places. Initially I suggested just to delete identical files, but after a closer look I realised that files are actually different.
For example we have:
bob/pad/face/config/database/msu_mfsd.py
and
bob/pad/face/config/msu_mfsd.py
Both are registered in setup.py, but in different groups.
One here:
``
'bob.pad.database': [
'msu-mfsd = bob.pad.face.config.database.msu_mfsd:database',
``
Another one here:
``
'bob.bio.config': [
'msu-mfsd = bob.pad.face.config.msu_mfsd',
``
As far as I understand first entry point allows to run experiment as follows:
``spoof.py --database msu-mfsd <other arguments>``
Second one is actually a full DB config allowing to run experiments easier:
``spoof.py msu-mfsd <other arguments>``
The question here is: should we try to get rid of one of these config files, or do we keep both?
I think one of these files can actually be removed, preserving functionality mentioned above.
@amohammadi Can you please advise on this? Thank you!Olegs NIKISINSOlegs NIKISINShttps://gitlab.idiap.ch/bob/bob.pad.face/-/issues/9Bob.bio.video is meant to be used as a wrapper class2018-03-09T11:25:59ZAmir MOHAMMADIBob.bio.video is meant to be used as a wrapper classWhen you have an extractor that works on one image like quality measures, you can use bob.bio.video to make it work on a video. This is explained in `bob.bio.video`'s documentation: https://www.idiap.ch/software/bob/docs/bob/bob.bio.vide...When you have an extractor that works on one image like quality measures, you can use bob.bio.video to make it work on a video. This is explained in `bob.bio.video`'s documentation: https://www.idiap.ch/software/bob/docs/bob/bob.bio.video/stable/implementation.html . Also @mguenther who have implemented this can explain this better.
But the way it is used in https://gitlab.idiap.ch/bob/bob.pad.face/blob/b0a14393f109e8bc15928ade60b0614e34e4b73f/bob/pad/face/extractor/VideoQualityMeasure.py#L25 is just a re-write of `bob.bio.video`.
This is true also for `VideoLBPHistogram` at least.Olegs NIKISINSOlegs NIKISINShttps://gitlab.idiap.ch/bob/docs/-/issues/2The math formulas are not rendered in the docs2018-03-09T14:08:33ZAmir MOHAMMADIThe math formulas are not rendered in the docsSee for example: https://www.idiap.ch/software/bob/docs/bob/docs/master/bob.measure/doc/guide.html
We probably are missing a sphinx extension in the config file.See for example: https://www.idiap.ch/software/bob/docs/bob/docs/master/bob.measure/doc/guide.html
We probably are missing a sphinx extension in the config file.https://gitlab.idiap.ch/bob/bob.bio.face/-/issues/28Cannot use FaceCrop with a video sequence where all frames do not have annota...2018-03-10T10:07:08ZAmir MOHAMMADICannot use FaceCrop with a video sequence where all frames do not have annotationsI have a video file with the missing annotations on some frames:
```
{
"0": {},
"1": {},
"2": {},
"3": {},
"4": {},
"5": {},
"6": {},
"7": {},
"8": {},
"9": {},
"10": {},
"11": {},
"12": {},
"13": {},
"14": {},
"15": {
...I have a video file with the missing annotations on some frames:
```
{
"0": {},
"1": {},
"2": {},
"3": {},
"4": {},
"5": {},
"6": {},
"7": {},
"8": {},
"9": {},
"10": {},
"11": {},
"12": {},
"13": {},
"14": {},
"15": {
"reye": [
635,
239
],
"leye": [
618,
647
],
}
```
and I want to use it FaceCrop wrapped in bob.bio.video with this video file. However it fails with:
```
bob.bio.base@2018-02-14 18:21:41,462 -- ERROR: During the execution, an exception was raised: At least one of the expected annotations '['leye', 'reye']' are not given in 'dict_keys([])'.
Traceback (most recent call last):
File "bin/verify.py", line 77, in <module>
sys.exit(bob.bio.base.script.verify.main())
File "src/bob.bio.base/bob/bio/base/script/verify.py", line 435, in main
verify(args, command_line_parameters)
File "src/bob.bio.base/bob/bio/base/script/verify.py", line 415, in verify
if not execute(args):
File "src/bob.bio.base/bob/bio/base/script/verify.py", line 264, in execute
force = args.force)
File "src/bob.bio.base/bob/bio/base/tools/preprocessor.py", line 81, in preprocess
preprocessed_data = preprocessor(data, annotations)
File "src/bob.bio.video/bob/bio/video/preprocessor/Wrapper.py", line 126, in __call__
preprocessed = self.preprocessor(frame, annots)
File "src/bob.bio.face/bob/bio/face/preprocessor/FaceCrop.py", line 208, in __call__
image = self.crop_face(image, annotations)
File "src/bob.bio.face/bob/bio/face/preprocessor/FaceCrop.py", line 154, in crop_face
raise ValueError("At least one of the expected annotations '%s' are not given in '%s'." % (self.cropped_keys, annotations.keys()))
ValueError: At least one of the expected annotations '['leye', 'reye']' are not given in 'dict_keys([])'.
```
What is the best strategy to avoid this error? I want the preprocessing to continue and drop those frames.