bob issueshttps://gitlab.idiap.ch/groups/bob/-/issues2017-11-04T00:38:17Zhttps://gitlab.idiap.ch/bob/bob.db.atvskeystroke/-/issues/1This database still uses bob.db.verification.utils2017-11-04T00:38:17ZAmir MOHAMMADIThis database still uses bob.db.verification.utilsI will remove this from bob.nightlies until this is fixedI will remove this from bob.nightlies until this is fixedRefactoring 2016 and gitlab migration milestoneAndré AnjosAndré Anjoshttps://gitlab.idiap.ch/bob/bob.bio.face/-/issues/25Reading original data from IJB-A2017-11-04T14:23:44ZTiago de Freitas PereiraReading original data from IJB-AHi,
It's been a long time that I don't run something on IJB-A and I'm struggling with something simple here.
The File wrapper in `bob.bio.base` reimplements the method `make_path` (https://gitlab.idiap.ch/bob/bob.bio.face/blob/d6f75c9a...Hi,
It's been a long time that I don't run something on IJB-A and I'm struggling with something simple here.
The File wrapper in `bob.bio.base` reimplements the method `make_path` (https://gitlab.idiap.ch/bob/bob.bio.face/blob/d6f75c9aaa58dc4fb6968a47aecced79b048419a/bob/bio/face/database/ijba.py#L22).
As you can see; the client id is appended to the path.
Although I understand the reason of doing that, the question is: How is this suppose to work? This clearly can't be used to read the original data.
Any clarification is appreciated
ThanksManuel Günthersiebenkopf@googlemail.comManuel Günthersiebenkopf@googlemail.comhttps://gitlab.idiap.ch/bob/bob.bio.base/-/issues/95A sequential and parallel preprocessor2017-11-06T15:28:33ZAmir MOHAMMADIA sequential and parallel preprocessorIt would be nice to have a helper class that takes several pre-processors and applies them to the data one by one sequentially.
Recently, I have been working with [madmom](https://madmom.readthedocs.io/en/latest/) and I really like the ...It would be nice to have a helper class that takes several pre-processors and applies them to the data one by one sequentially.
Recently, I have been working with [madmom](https://madmom.readthedocs.io/en/latest/) and I really like the idea of [functions/classes API](https://github.com/CPJKU/madmom_tutorials/blob/master/audio_signal_handling.ipynb) they provide + their [processors API](https://github.com/CPJKU/madmom_tutorials/blob/master/processors.ipynb). We could have something similar throughout Bob generally I think.https://gitlab.idiap.ch/bob/bob.bio.base/-/issues/67ZT files are processed even when no ZT processing is wanted2017-11-10T17:03:20ZManuel Günthersiebenkopf@googlemail.comZT files are processed even when no ZT processing is wantedWhen a ZT database is instantiated, but used without ZT normalization enabled, the files for ZT norm are still processed. This is annoying.
A fix would include changing the basic database interface of the `all_files` function: https://g...When a ZT database is instantiated, but used without ZT normalization enabled, the files for ZT norm are still processed. This is annoying.
A fix would include changing the basic database interface of the `all_files` function: https://gitlab.idiap.ch/bob/bob.bio.base/blob/master/bob/bio/base/database/database.py#L370 and https://gitlab.idiap.ch/bob/bob.bio.base/blob/master/bob/bio/base/database/database.py#L640
The fix would introduce a new parameter, which tells if ZT score files should be included or not.
When called from the FileSelector, e.g.: https://gitlab.idiap.ch/bob/bob.bio.base/blob/master/bob/bio/base/tools/FileSelector.py#L125 The information need to be provided.
How this is done, I need to investigate.Manuel Günthersiebenkopf@googlemail.comManuel Günthersiebenkopf@googlemail.comhttps://gitlab.idiap.ch/bob/bob.learn.mlp/-/issues/9unchecked functions from `bob::math` are used2017-11-14T05:01:03ZManuel Günthersiebenkopf@googlemail.comunchecked functions from `bob::math` are usedsee bob/bob.math!13see bob/bob.math!13Manuel Günthersiebenkopf@googlemail.comManuel Günthersiebenkopf@googlemail.comhttps://gitlab.idiap.ch/bob/bob.learn.mlp/-/issues/8constness issue in nose tests2017-11-14T05:01:03ZAndré Anjosconstness issue in nose tests*Created by: siebenkopf*
With the new constness initiative (see: https://github.com/bioidiap/bob.blitz/issues/6) there are some nose tests failing:
https://travis-ci.org/bioidiap/bob.learn.mlp/builds/120843870
as they try to modify va...*Created by: siebenkopf*
With the new constness initiative (see: https://github.com/bioidiap/bob.blitz/issues/6) there are some nose tests failing:
https://travis-ci.org/bioidiap/bob.learn.mlp/builds/120843870
as they try to modify values that are ``const``. Please review both the tests and also whether these values must be returned as ``const`` arrays.https://gitlab.idiap.ch/bob/bob.learn.mlp/-/issues/5bob.learn.mlp.Machine.__repr__ crashes2017-11-14T05:01:03ZAndré Anjosbob.learn.mlp.Machine.__repr__ crashes*Created by: siebenkopf*
The ``__repr__`` method of ``bob.learn.mlp.Machine`` crashes, throughout. Try:
```python
>>> import bob.learn.mlp
>>> m = bob.learn.mlp.Machine((1,1,1))
>>> print (m)
```
The reason is that the ``PyBob...*Created by: siebenkopf*
The ``__repr__`` method of ``bob.learn.mlp.Machine`` crashes, throughout. Try:
```python
>>> import bob.learn.mlp
>>> m = bob.learn.mlp.Machine((1,1,1))
>>> print (m)
```
The reason is that the ``PyBobLearnMLPMachine_Repr`` method:
https://github.com/bioidiap/bob.learn.mlp/blob/master/bob/learn/mlp/machine.cpp#L702
expects the ``PyBobLearnMLPMachine_getWeights`` to return a ``PyBlitzArray_Object*``, while it actually returns a tuple:
https://github.com/bioidiap/bob.learn.mlp/blob/master/bob/learn/mlp/machine.cpp#L261
Hence, ``NULL`` is passed to ``make_safe``, which crashes with a SEGFAULT.https://gitlab.idiap.ch/bob/bob.learn.mlp/-/issues/3Setting input_subtract and input_divide in bob.learn.mlp.Machine2017-11-14T05:01:03ZAndré AnjosSetting input_subtract and input_divide in bob.learn.mlp.Machine*Created by: ivana7c*
The PyNumber_Check seems to pass also for numpy.array, and returns an error when you try to use this setter with numpy.array:
https://github.com/bioidiap/bob.learn.mlp/blob/master/bob/learn/mlp/machine.cpp#L438
...*Created by: ivana7c*
The PyNumber_Check seems to pass also for numpy.array, and returns an error when you try to use this setter with numpy.array:
https://github.com/bioidiap/bob.learn.mlp/blob/master/bob/learn/mlp/machine.cpp#L438
The same applies for the input_divide setter.https://gitlab.idiap.ch/bob/bob.learn.mlp/-/issues/2Where is this module install to?2017-11-14T05:01:03ZAndré AnjosWhere is this module install to?*Created by: StevenLOL*
Where is this module install to?
Follow the readme:
$ python bootstrap.py
...
$ ./bin/buildout
And yet unalbe to find where it install to. Please help.
An so this is the error:
mlp = bob....*Created by: StevenLOL*
Where is this module install to?
Follow the readme:
$ python bootstrap.py
...
$ ./bin/buildout
And yet unalbe to find where it install to. Please help.
An so this is the error:
mlp = bob.learn.mlp.Machine((3, 3, 2, 1))
AttributeError: 'module' object has no attribute 'learn'
https://gitlab.idiap.ch/bob/bob.learn.mlp/-/issues/1Travis builds are not passing2017-11-14T05:01:03ZAndré AnjosTravis builds are not passing*Created by: siebenkopf*
Hi,
currently the Travis builds of this package are not passing:
https://travis-ci.org/bioidiap/bob.learn.mlp
It seems that there is an issue with the doctests. Unfortunately, I cannot reproduce this issue ...*Created by: siebenkopf*
Hi,
currently the Travis builds of this package are not passing:
https://travis-ci.org/bioidiap/bob.learn.mlp
It seems that there is an issue with the doctests. Unfortunately, I cannot reproduce this issue on my machine.
Please investigate!https://gitlab.idiap.ch/bob/bob.db.asvspoof/-/issues/4add user `biometrics` to pypi2017-11-15T05:40:21ZTiago de Freitas Pereiraadd user `biometrics` to pypiHi,
I wasn't able to publish this package, because the user `biometrics` is not allowed. Neither am I.
Could you please do it @pkorshunov?
check the log https://gitlab.idiap.ch/bob/bob.db.asvspoof/builds/19143
Once done, just rer...Hi,
I wasn't able to publish this package, because the user `biometrics` is not allowed. Neither am I.
Could you please do it @pkorshunov?
check the log https://gitlab.idiap.ch/bob/bob.db.asvspoof/builds/19143
Once done, just rerun the stage that failed here.
https://gitlab.idiap.ch/bob/bob.db.asvspoof/pipelines/4427#deploy
Thanks
Pavel KORSHUNOVPavel KORSHUNOVhttps://gitlab.idiap.ch/bob/bob.db.asvspoof/-/issues/3RuntimeError: Database 'asvspoof' cannot be found at expected location2017-11-15T05:40:21ZAmir MOHAMMADIRuntimeError: Database 'asvspoof' cannot be found at expected location```
ERROR: bob.bio.db.tests.test_implementations.test_ASVspoofBioDatabase
----------------------------------------------------------------------
Traceback (most recent call last):
File "/local/gitlab-runner/builds/224a6381/0/bob/bo...```
ERROR: bob.bio.db.tests.test_implementations.test_ASVspoofBioDatabase
----------------------------------------------------------------------
Traceback (most recent call last):
File "/local/gitlab-runner/builds/224a6381/0/bob/bob.nightlies/env/lib/python2.7/site-packages/nose/case.py", line 197, in runTest
self.test(*self.arg)
File "/local/gitlab-runner/builds/224a6381/0/bob/bob.nightlies/env/lib/python2.7/site-packages/bob/bio/db/tests/test_implementations.py", line 77, in test_ASVspoofBioDatabase
_check_database(database, training_depends=True, skip_train=True)
File "/local/gitlab-runner/builds/224a6381/0/bob/bob.nightlies/env/lib/python2.7/site-packages/bob/bio/db/tests/test_implementations.py", line 38, in _check_database
assert len(database.all_files()) > 0
File "/local/gitlab-runner/builds/224a6381/0/bob/bob.nightlies/env/lib/python2.7/site-packages/bob/bio/db/database.py", line 449, in all_files
return self.sort(self.objects(protocol=self.protocol, groups=groups, **self.all_files_options))
File "/local/gitlab-runner/builds/224a6381/0/bob/bob.nightlies/env/lib/python2.7/site-packages/bob/bio/db/databases/asvspoof.py", line 121, in objects
clients=model_ids, gender=gender)
File "/local/gitlab-runner/builds/224a6381/0/bob/bob.nightlies/env/lib/python2.7/site-packages/bob/db/asvspoof/query.py", line 106, in objects
self.assert_validity()
File "/local/gitlab-runner/builds/224a6381/0/bob/bob.nightlies/env/lib/python2.7/site-packages/bob/db/asvspoof/query.py", line 64, in assert_validity
INFO.name(), SQLITE_FILE))
RuntimeError: Database 'asvspoof' cannot be found at expected location '/local/gitlab-runner/builds/224a6381/0/bob/bob.nightlies/env/lib/python2.7/site-packages/bob/db/asvspoof/db.sql3'. Create it and then try re-connecting using Database.connect()
```
I see that this is being downloaded in the nightlies:
```
$ ./bin/bob_dbmanage.py all download
Extracting url 'https://www.idiap.ch/software/bob/databases/latest/asvspoof.tar.bz2' to '/local/gitlab-runner/builds/224a6381/0/bob/bob.nightlies/src/bob.db.asvspoof/bob/db/asvspoof/db.sql3'
```
But the `db.sql3` in the wheel. While all other databases including `avspoof` have their `db.sql3` in their wheel.Refactoring 2016 and gitlab migration milestonePavel KORSHUNOVPavel KORSHUNOVhttps://gitlab.idiap.ch/bob/bob.db.asvspoof/-/issues/2Seems encoding is not set to unicode in macosx CI2017-11-15T05:40:21ZPavel KORSHUNOVSeems encoding is not set to unicode in macosx CIBuilding on macosx_34 is failing due to encoding problem when reading README.rts.
Here is the error:
UnicodeDecodeError: 'ascii' codec can't decode byte 0xef in position 1454: ordinal not in range(128)
The build with the error is ...Building on macosx_34 is failing due to encoding problem when reading README.rts.
Here is the error:
UnicodeDecodeError: 'ascii' codec can't decode byte 0xef in position 1454: ordinal not in range(128)
The build with the error is here:
https://gitlab.idiap.ch/bob/bob.db.asvspoof/builds/9164Amir MOHAMMADIAmir MOHAMMADIhttps://gitlab.idiap.ch/bob/bob.db.asvspoof/-/issues/1Please drop support for bob.db.verification.utils and antispoofing.utils2017-11-15T05:40:21ZAndré AnjosPlease drop support for bob.db.verification.utils and antispoofing.utils*Created by: 183amir*
*Created by: 183amir*
Pavel KORSHUNOVPavel KORSHUNOVhttps://gitlab.idiap.ch/bob/bob.pad.face/-/issues/10High-level db interfaces here are not compatible with bob.bio.video2017-11-24T09:05:51ZAmir MOHAMMADIHigh-level db interfaces here are not compatible with bob.bio.videoWhile database interfaces here return frame containers just like databases in `bob.bio.video`, the API is incompatible with bob.bio.video and the frame selector is hard coded inside database interfaces. For an example, see: https://gitla...While database interfaces here return frame containers just like databases in `bob.bio.video`, the API is incompatible with bob.bio.video and the frame selector is hard coded inside database interfaces. For an example, see: https://gitlab.idiap.ch/bob/bob.pad.face/blob/b0a14393f109e8bc15928ade60b0614e34e4b73f/bob/pad/face/database/replay.py#L50 especially line 71.
The way it is written now, not only you cannot use bob.bio.video wrapper classes with the databases here but also they always return all the frames and leave no option to the user.
Again @mguenther would be the best person to explain this.Amir MOHAMMADIAmir MOHAMMADIhttps://gitlab.idiap.ch/bob/bob.db.oulunpu/-/issues/1Nightlies2017-11-28T11:58:09ZTiago de Freitas PereiraNightliesCould someone please have a look at it?
https://gitlab.idiap.ch/bob/bob.nightlies/-/jobs/106295
https://gitlab.idiap.ch/bob/bob.nightlies/-/jobs/106298
ThanksCould someone please have a look at it?
https://gitlab.idiap.ch/bob/bob.nightlies/-/jobs/106295
https://gitlab.idiap.ch/bob/bob.nightlies/-/jobs/106298
Thankshttps://gitlab.idiap.ch/bob/bob.pad.face/-/issues/16Nightlies2017-11-28T12:00:21ZTiago de Freitas PereiraNightliesIt seems a consistent issue along the builds.
Could someone please have a look at it?
https://gitlab.idiap.ch/bob/bob.nightlies/-/jobs/107120
https://gitlab.idiap.ch/bob/bob.nightlies/-/jobs/107123
https://gitlab.idiap.ch/bob/bob.nigh...It seems a consistent issue along the builds.
Could someone please have a look at it?
https://gitlab.idiap.ch/bob/bob.nightlies/-/jobs/107120
https://gitlab.idiap.ch/bob/bob.nightlies/-/jobs/107123
https://gitlab.idiap.ch/bob/bob.nightlies/-/jobs/107122
https://gitlab.idiap.ch/bob/bob.nightlies/-/jobs/107121
Thankshttps://gitlab.idiap.ch/bob/bob.extension/-/issues/46Name clashes when loading several run-time config files.2017-11-30T14:41:08ZAmir MOHAMMADIName clashes when loading several run-time config files.I have noticed the loaded config files are loaded into a module called `config` in [here](https://gitlab.idiap.ch/bob/bob.extension/blob/820afbee3cecf766c1b0b23244be89d88d911168/bob/extension/config.py#L72).
but in bob.bio.base this name...I have noticed the loaded config files are loaded into a module called `config` in [here](https://gitlab.idiap.ch/bob/bob.extension/blob/820afbee3cecf766c1b0b23244be89d88d911168/bob/extension/config.py#L72).
but in bob.bio.base this name is selected randomly each time.
Shouldn't this be like bob.bio.base as well?https://gitlab.idiap.ch/bob/bob.bio.face/-/issues/27Customized tables2017-12-01T12:53:18ZTiago de Freitas PereiraCustomized tablesWould be very practical to have a script similar to `bob.bio.base.script.collect_results` that dumps very nice copy/paste rst table like this one (for the verification protocols):
```
+-----------+-------------- +----------------+------...Would be very practical to have a script similar to `bob.bio.base.script.collect_results` that dumps very nice copy/paste rst table like this one (for the verification protocols):
```
+-----------+-------------- +----------------+-----------------+------------------------------------------+
| CMC(R=1) | FRR (FAR=0.1) | FRR (FAR=0.01) | FRR (FAR=0.001) | directory |
+-----------+---------------+----------------+-----------------+------------------------------------------+
| 96.062% | 12.035% | 42.485% | 74.931% | /vgg16/compare/split1/compare_split1 |
+-----------+---------------+----------------+-----------------+------------------------------------------+
| 96.753% | 11.580% | 36.706% | 63.294% | /vgg16/compare/split2/compare_split2 |
+-----------+---------------+----------------+-----------------+------------------------------------------+
| 95.778% | 11.556% | 47.038% | 69.803% | /vgg16/compare/split3/compare_split3 |
+-----------+---------------+----------------+-----------------+------------------------------------------+
| 97.830% | 11.185% | 36.672% | 62.883% | /vgg16/compare/split4/compare_split4 |
+-----------+---------------+----------------+-----------------+------------------------------------------+
| 95.838% | 12.315% | 43.330% | 66.762% | /vgg16/compare/split5/compare_split5 |
+-----------+---------------+----------------+-----------------+------------------------------------------+
| 95.449% | 11.650% | 45.752% | 72.573% | /vgg16/compare/split6/compare_split6 |
+-----------+---------------+----------------+-----------------+------------------------------------------+
| 96.429% | 10.291% | 41.344% | 64.891% | /vgg16/compare/split7/compare_split7 |
+-----------+---------------+----------------+-----------------+------------------------------------------+
| 95.891% | 10.459% | 36.660% | 65.422% | /vgg16/compare/split8/compare_split8 |
+-----------+---------------+----------------+-----------------+------------------------------------------+
| 95.635% | 14.551% | 43.933% | 69.045% | /vgg16/compare/split9/compare_split9 |
+-----------+---------------+----------------+-----------------+------------------------------------------+
| 96.062% | 12.035% | 42.485% | 74.931% | /vgg16/compare/split10/compare_split10 |
+-----------+---------------+----------------+-----------------+------------------------------------------+
```https://gitlab.idiap.ch/bob/bob.bio.base/-/issues/98Scoring phase is too non-verbose even when `-vvv` is passed2017-12-05T10:25:54ZAndré AnjosScoring phase is too non-verbose even when `-vvv` is passedWhile executing `verify.py`, the preprocessing and extraction of each sample is carefully reported when in `-vvv` mode. The same does not happen when scoring. There is absolutely no output.
It would be good if the scoring bit output the...While executing `verify.py`, the preprocessing and extraction of each sample is carefully reported when in `-vvv` mode. The same does not happen when scoring. There is absolutely no output.
It would be good if the scoring bit output the number of models to be scored and, as it advances from one model to another, a printout of the model-identifier that was just scored.
@mguenther: is that doable?Manuel Günthersiebenkopf@googlemail.comManuel Günthersiebenkopf@googlemail.com