Add store load functionality #52
Conversation
brian2modelfitting/inferencer.py
Outdated
| t = self.theta | ||
| else: | ||
| raise AttributeError('Provide sampled prior or call ' | ||
| '``infere_step``method first.') |
There was a problem hiding this comment.
tiny detail: I wouldn't put the restructured text formating with double backquotes into the error message (and there's a space missing).
Same thing in several other error messages.
There was a problem hiding this comment.
Should I do
raise AttributeError('Provide sampled prior or call '
'`infere_step` method first.')or something like
raise AttributeError("Provide sampled prior or call "
"'infere_step' method first.")instead?
brian2modelfitting/inferencer.py
Outdated
| Consisting of sampled prior and summary statistics arrays. | ||
| """ | ||
| loaded = np.load(f, allow_pickle=True) | ||
| if loaded.files == ['theta', 'x'] or loaded.files == ['x', 'theta']: |
There was a problem hiding this comment.
if set(loaded.files) == {'theta', 'x'} maybe?
brian2modelfitting/inferencer.py
Outdated
| if loaded.files == ['theta', 'x'] or loaded.files == ['x', 'theta']: | ||
| theta = loaded['theta'] | ||
| x = loaded['x'] | ||
| elif loaded.files == ['arr_0', 'arr_1']: |
There was a problem hiding this comment.
Do we need to support this? Who would be storing things as arr_0, arr_1?
There was a problem hiding this comment.
Who would be storing things as arr_0, arr_1?
A crazy person 😄
Kidding, if you do not provide the keyword arguments when you store compressed arrays, numpy will name the arrays arr_0, arr_1, ... by default. This will happen only if the user stores training data manually, which should not be the case because there is a method already implemented for that. So, yeah, you are right. We do not need to support that and I will remove that now.
examples/hh_sbi_advanced.py
Outdated
| r'$\overline{C}_{m}$']) | ||
| # ...and optionally, continue the multiround inference via ``infere`` method | ||
| posterior_multi_round = inferencer.infere(n_rounds=2, | ||
| theta=theta_loaded, x=x_loaded, |
There was a problem hiding this comment.
Shouldn't this be based on the sampled posterior instead of putting in the sampled prior data again (theta_loaded / x_loaded)?
There was a problem hiding this comment.
Yes, my mistake. Nothing will happen by specifying theta and x because they are already defined inside the class. Posterior is also stored in a class variable, so you should only specify n_rounds here.
Resolving #46.