Thomas Moreau

I asked @BadrAlpha07 to open this issue because several students had issues understanding what was the problem with there submissions for classifiers that where not detected as such. I think...

There is a way to check if a call is in cache with [`check_call_in_cache`](https://joblib.readthedocs.io/en/latest/memory.html#joblib.memory.MemorizedFunc.check_call_in_cache). The workflow is more like: ```python func_cache = memory.cache(func) hit = func_cache.check_call_in_cache(*args, **kwargs) ``` It is...

Yes, that's true it requires computing the hash twice. If there is use-cases where avoiding this is necessary, I would rather add a function `call_with_is_cached` that returns the both the...

Can't you do it in the `get_data` method of the `Dataset`? this should only gets called once.

Then, I think we should have an extra method for this. Something like `setup_data`, that gets called after `set_data` only if it has not been called before or the data...

Just got the same issue, the proposed fix seems nice and adding an option to tweak it if necessary would probably also be useful. Thanks a lot for maintaining and...

It depends what you mean by safe :slightly_smiling_face:. If you mean safe to concurrent write, I think `joblib` is safe at least for multithreading and multiprocessing, as it first write...

I think so. As in python3, we rely on the `__cause__` attribute of an exception to report the true error to the user, we should make sure it exists.

> @tomMoral I've made your requested changes (including changing `age_limit` to a timedelta; not sure why I can't reply to that comment). Thanks a lot for the modifications. > I...

Hi, thanks for reporting. This is actually the expected behavior for `loky`. The `loky` backend rely on `spawn` to start the new processes for the pool of worker. As this...