How to use callbacks

[dmpetrov]

I'm creating a callback "model":

>>> def pre(x: list) -> float: return x[0]*x[1]*x[2]*x[3]
>>> m = save(pre, "my_func1")

However, it cannot be applied to a dataframe d1-m:

$ mlem apply my_func d1-m --output of
⏳️ Loading data from d1-m.mlem
⏳️ Loading model from my_func.mlem
🍏 Applying `predict` method...
❌ Unexpected error: 'predict'
Use the --tb or --traceback option to include the traceback in the output
Please report it here: <https://github.com/iterative/mlem/issues>

How to do that properly?

[aguschin]

@dmpetrov, it's

$ mlem apply my_func1 d1-m --output of --method __call__
⏳️ Loading data from d1-m.mlem
⏳️ Loading model from my_func1.mlem
🍏 Applying `__call__` method...
💾 Saving data to of.mlem

MLEM should use __call__ here by default.