[livdet-iris-2020] Fix category type in view

The view expects an int32 for the category output.

@samuel.gaist I realized the database should be called livdet-iris-2020 and not livedet... is it also possible to rename it?

added 1 commit

• beb18f4d - [livdet-iris-2020] rename the db and a small documentation

Compare with previous version

changed title from [livedet-iris-2020] Fix category type in view to [livdet-iris-2020] Fix category type in view

We have an interesting situation here:

• some experiments are likely already using it
• This is technically a correction so it should be version 2
• But since you need to rename that database it should be considered a new one and stay at 1

What is your take here @andre.anjos ?

You can delete the old db and associated experiments. No one (me) would mind ;)

@amohammadi: you should delete the old experiments yourself.

Then, we can update the view. If that is not possible technically, then we have either to change the name of the database or bump the version.

@amohammadi: you should delete the old experiments yourself.

done.

btw the error for this was like this:

$ beat exp run amohammadi/amohammadi/livdet/1/livedet
Skipping execution of `amohammadi/livdet_baseline/1' for block `algorithm' - outputs exist
Running `amohammadi/livdet_analyzer/1' for block `analyzer'
Start the execution of 'amohammadi/livdet_analyzer/1'
Block did not execute properly - outputs were reset
  Standard output:
    
  Standard error:
    
  Captured user error:
    File "/home/amir/miniconda/envs/iris/lib/python3.7/site-packages/beat/backend/python/baseformat.py", line 115, in setup_scalar
        % (attrname, formatname, numpy.array(value).dtype, dtype)
    TypeError: cannot safely cast attribute `value' on dataformat `system/integer/1' with type `int64' to `int32' without precision loss
  Captured system error:
    
Error: Error occured: returned value is 1
Removing cache files: No data written

which was very hard to debug.

If it's not already in the documentation, I at least explained when presenting the platform that it's important to properly type what you write to the outputs as type conversion may fail for values as simple as 0.

I know that values need to be casted. The problem here is that the backtrace error I was getting was not telling me this error was in the view and not in my analyzer (the step I was on when hitting this error). So, I kept looking at my analyzer and thinking why I get this error while the error was in the view.

I see where the issue might be here: the logs of both containers are concatenated without context.

Can you open an issue on beat/beat.core> to improve that.

merged

mentioned in commit 5265c226

mentioned in issue beat.core#100