Created a dummy db that dump some fake missing files

Hi,

In order to test this MR bob.bio.gmm!13 (merged) , I needed to create a dummy database that dumps a "missing" file.

I will reassign this to @mguenther, in case we merge this without his eyes on it (thanks to the time zone)

assigned to @mguenther

Why are you implementing a new database for this. This option is already available, and it is tested here: https://gitlab.idiap.ch/bob/bob.bio.base/blob/700d7a48ac8f58f59dc5a74ce1e1662d21bdc795/bob/bio/base/test/test_scripts.py#L289 As you can see, the DummyDatabase already gets an option to return None, throughout, which means that all of the original data is marked as missing. Maybe you can change this option (or add another option) to that database so that you don't need to implement yet another dummy database.

Ok, I patch the preprocessor and add a keywork called none_probability that will dump None at certain probability.

class DummyPreprocessor (Preprocessor):
  def __init__(self, return_none=False, none_probability=1, **kwargs):

added 1 commit

• 8129781d - Created a dummy db that dump some fake missing files

Compare with previous version

added 1 commit

• e6983a19 - Removed allow-missing files in the raw data reading

Compare with previous version

Done, I'm implemented the thing in the DummyPreprocessor.

With this merged, I will move on to the original issue in bob.bio.gmm.

added 1 commit

• 51548aa8 - removed space

Compare with previous version

added 1 commit

• 53cd7d12 - Aligned spaces

Compare with previous version

resolved all discussions

I'll set it to auto-merge. Thanks @tiago.pereira

enabled an automatic merge when the pipeline for 53cd7d12 succeeds

merged

mentioned in commit cec5e5cb