Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fix modelbuilder load #229

Open
wants to merge 10 commits into
base: main
Choose a base branch
from
+618 −44
Open

Fix modelbuilder load #229

Show file tree
Hide file tree
Changes from 8 commits
Commits
Show all changes
10 commits
Select commit Hold shift + click to select a range
e553f47
Updated fit and predict docstring
Aug 15, 2023
1aa690b
Rework of modelbuilder save, load, fit, and predict
Aug 15, 2023
57ccdb8
Updated the example notebook
Aug 15, 2023
c89022b
Update modelbuilder_example.ipynb
Aug 15, 2023
f6cdcca
Update modelbuilder_example.ipynb
Aug 15, 2023
0a6f1dc
Update modelbuilder_example.ipynb
Aug 15, 2023
37c4161
Rework of modelbuilder save, load, fit and predict
Aug 15, 2023
c2ba4bd
Merge branch 'fix_modelbuilder_load' of https://github.com/AuguB/pymc…
Aug 15, 2023
1198415
Revised modelbuilder
Aug 16, 2023
4370eca
Added my name to the top of the file
AuguB Oct 11, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
541 changes: 541 additions & 0 deletions notebooks/modelbuilder_example.ipynb
View file
Open in desktop

Large diffs are not rendered by default.

86 changes: 48 additions & 38 deletions pymc_experimental/model_builder.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -74,8 +74,8 @@ def __init__(
sampler_config = self.default_sampler_config if sampler_config is None else sampler_config
self.sampler_config = sampler_config
model_config = self.default_model_config if model_config is None else model_config

self.model_config = model_config # parameters for priors etc.
self.model_coords = None
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think this property should be included in child classes IF they utilise coords. Otherwise it forces all classes that would inherit from MB even if they don't have Hierarchical nature.

Copy link
Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks for your suggestion. The suggestion does not force any non-hierarchical model to populate the coords attribute. If a model has no coords, this can stay None

self.model = None # Set by build_model
self.idata: Optional[az.InferenceData] = None # idata is generated during fitting
self.is_fitted_ = False
Expand Down Expand Up @@ -172,7 +172,7 @@ def default_sampler_config(self) -> Dict:
--------
>>> @classmethod
>>> def default_sampler_config(self):
>>> Return {
>>> return {
>>> 'draws': 1_000,
>>> 'tune': 1_000,
>>> 'chains': 1,
Expand All @@ -187,13 +187,10 @@ def default_sampler_config(self) -> Dict:
raise NotImplementedError

@abstractmethod
def generate_and_preprocess_model_data(
self, X: Union[pd.DataFrame, pd.Series], y: pd.Series
) -> None:
def preprocess_model_data(self, X: Union[pd.DataFrame, pd.Series], y: pd.Series = None) -> None:
Copy link
Collaborator

@michaelraczycki michaelraczycki Aug 15, 2023 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Signature doesn't match the return value from docstring, will result in mypy error in child class about breaking Liskov's principle if it was implemented in a way suggested in a docstring

"""
Applies preprocessing to the data before fitting the model.
if validate is True, it will check if the data is valid for the model.
sets self.model_coords based on provided dataset

Parameters:
X : array, shape (n_obs, n_features)
Expand All @@ -202,17 +199,16 @@ def generate_and_preprocess_model_data(
Examples
--------
>>> @classmethod
>>> def generate_and_preprocess_model_data(self, X, y):
>>> def preprocess_model_data(self, X, y):
>>> x = np.linspace(start=1, stop=50, num=100)
>>> y = 5 * x + 3 + np.random.normal(0, 1, len(x)) * np.random.rand(100)*10 + np.random.rand(100)*6.4
>>> X = pd.DataFrame(x, columns=['x'])
>>> y = pd.Series(y, name='y')
>>> self.X = X
>>> self.y = y
>>> return X, y

Returns
-------
None
pd.DataFrame, pd.Series

"""
raise NotImplementedError
Expand Down Expand Up @@ -258,6 +254,23 @@ def build_model(
"""
raise NotImplementedError

def save_model_coords(self, X: Union[pd.DataFrame, pd.Series], y: pd.Series):
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This should be an abstract method, since it doesn't actually implement anything (the assignment of None, to a property which is originally initialised to None)

Copy link
Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yes, this method does not do anything. I don't know exactly how python abstract methods work, but like this I hoped the specific implementation would be optional for those cases where the model coords actually need to be set. For non-hierarchical models, this method does not need to be touched.

Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The reason I didn't even include it here is exactly what you're mentioning here: A lot of models won't use the coords at all, but this here would force the models to contain property that will never be used, along with its setter method. In my opinion is best to not include it at all, and allow for developers of specific child model to include this, as it was done in case of DelayedSaturatedMMM from pymc-marketing

Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Why does save_model_coords require both X and y as inputs?

Copy link
Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I did that because I thought there may be some highly unusual cases in which y is also required. I would have no problem with it myself if this function only takes X.

Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

In that case y should be optional param, with default value being None

"""Creates the model coords.

Parameters:
X : array, shape (n_obs, n_features)
y : array, shape (n_obs,)

Examples
--------
def set_model_coords(self, X, y):
group_dim1 = X['group1'].unique()
group_dim2 = X['group2'].unique()

self.model_coords = {'group1':group_dim1, 'group2':group_dim2}
"""
self.model_coords = None

def sample_model(self, **kwargs):
"""
Sample from the PyMC model.
Expand Down Expand Up @@ -339,6 +352,7 @@ def set_idata_attrs(self, idata=None):
idata.attrs["version"] = self.version
idata.attrs["sampler_config"] = json.dumps(self.sampler_config)
idata.attrs["model_config"] = json.dumps(self._serializable_model_config)
idata.attrs["model_coords"] = json.dumps(self.model_coords)
# Only classes with non-dataset parameters will implement save_input_params
if hasattr(self, "_save_input_params"):
self._save_input_params(idata)
Expand Down Expand Up @@ -432,18 +446,12 @@ def load(cls, fname: str):
model_config=model_config,
sampler_config=json.loads(idata.attrs["sampler_config"]),
)
model.model_coords = json.loads(idata.attrs["model_coords"])
model.idata = idata
dataset = idata.fit_data.to_dataframe()
X = dataset.drop(columns=[model.output_var])
y = dataset[model.output_var]
model.build_model(X, y)
# All previously used data is in idata.

if model.id != idata.attrs["id"]:
raise ValueError(
f"The file '{fname}' does not contain an inference data of the same model or configuration as '{cls._model_type}'"
)

return model

def fit(
Expand All @@ -462,7 +470,7 @@ def fit(

Parameters
----------
X : array-like if sklearn is available, otherwise array, shape (n_obs, n_features)
X : pd.DataFrame (n_obs, n_features)
The training input samples.
y : array-like if sklearn is available, otherwise array, shape (n_obs,)
The target values (real numbers).
Expand Down Expand Up @@ -492,26 +500,15 @@ def fit(
if y is None:
y = np.zeros(X.shape[0])
y = pd.DataFrame({self.output_var: y})
self.generate_and_preprocess_model_data(X, y.values.flatten())
self.build_model(self.X, self.y)
X_prep, y_prep = self.preprocess_model_data(X, y.values.flatten())
self.save_model_coords(X_prep, y_prep)
self.build_model(X_prep, y_prep)

sampler_config = self.sampler_config.copy()
sampler_config["progressbar"] = progressbar
sampler_config["random_seed"] = random_seed
sampler_config.update(**kwargs)
self.idata = self.sample_model(**sampler_config)

X_df = pd.DataFrame(X, columns=X.columns)
combined_data = pd.concat([X_df, y], axis=1)
assert all(combined_data.columns), "All columns must have non-empty names"
with warnings.catch_warnings():
warnings.filterwarnings(
"ignore",
category=UserWarning,
message="The group fit_data is not defined in the InferenceData scheme",
)
self.idata.add_groups(fit_data=combined_data.to_xarray()) # type: ignore

return self.idata # type: ignore

def predict(
Expand All @@ -526,7 +523,7 @@ def predict(

Parameters
---------
X_pred : array-like if sklearn is available, otherwise array, shape (n_pred, n_features)
X_pred : pd.DataFrame (n_pred, n_features)
The input data used for prediction.
extend_idata : Boolean determining whether the predictions should be added to inference data object.
Defaults to True.
Expand All @@ -545,9 +542,12 @@ def predict(
>>> prediction_data = pd.DataFrame({'input':x_pred})
>>> pred_mean = model.predict(prediction_data)
"""

synth_y = pd.Series(np.zeros(len(X_pred)))
X_pred_prep, y_synth_prep = self.preprocess_model_data(X_pred, synth_y)
if self.model is None:
self.build_model(X_pred_prep, y_synth_prep)
posterior_predictive_samples = self.sample_posterior_predictive(
X_pred, extend_idata, combined=False, **kwargs
X_pred_prep, extend_idata, combined=False, **kwargs
)

if self.output_var not in posterior_predictive_samples:
Expand Down Expand Up @@ -652,6 +652,7 @@ def get_params(self, deep=True):
return {
"model_config": self.model_config,
"sampler_config": self.sampler_config,
"model_coords": self.model_coords,
}

def set_params(self, **params):
Expand All @@ -660,6 +661,7 @@ def set_params(self, **params):
"""
self.model_config = params["model_config"]
self.sampler_config = params["sampler_config"]
self.model_coords = params["model_coords"]

@property
@abstractmethod
Expand All @@ -682,7 +684,11 @@ def predict_proba(
**kwargs,
) -> xr.DataArray:
"""Alias for `predict_posterior`, for consistency with scikit-learn probabilistic estimators."""
return self.predict_posterior(X_pred, extend_idata, combined, **kwargs)
synth_y = pd.Series(np.zeros(len(X_pred)))
X_pred_prep, y_synth_prep = self.preprocess_model_data(X_pred, synth_y)
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

preprocess_model_data should not be defined in a way that forces creation of entire dummy dataset for y if we want to process X, so: preprocess_model_data(X_pred: pd.DataFrame, y_pred:Optional[pd.Series]= None) and the logic that checks first if the y was provided.

if self.model is None:
self.build_model(X_pred_prep, y_synth_prep)
return self.predict_posterior(X_pred_prep, extend_idata, combined, **kwargs)

def predict_posterior(
self,
Expand Down Expand Up @@ -710,9 +716,13 @@ def predict_posterior(
Posterior predictive samples for each input X_pred
"""

X_pred = self._validate_data(X_pred)
synth_y = pd.Series(np.zeros(len(X_pred)))
X_pred_prep, y_synth_prep = self.preprocess_model_data(X_pred, synth_y)
if self.model is None:
self.build_model(X_pred_prep, y_synth_prep)

posterior_predictive_samples = self.sample_posterior_predictive(
X_pred, extend_idata, combined, **kwargs
X_pred_prep, extend_idata, combined=False, **kwargs
)

if self.output_var not in posterior_predictive_samples:
Expand Down