Skip to content

Add ty type checker and fix errors #114

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

Open
elbersb wants to merge 18 commits into
base: master
Choose a base branch
from
Open

Add ty type checker and fix errors #114

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
18 commits
Select commit Hold shift + click to select a range
2596ac2
Add ty
elbersb Jan 13, 2026
38c2b4f
Two lines added: @abstractmethod decorators for methods that already …
elbersb Jan 13, 2026
feea925
Add Optional for types that can be None
elbersb Jan 13, 2026
4d71acb
Continued fixing invalid-parameter-default errors by adding Optional[…
elbersb Jan 13, 2026
597f784
Remove dead code: unused functions, constants, and commented stubs
elbersb Jan 13, 2026
7694213
Fix more return types
elbersb Jan 13, 2026
d80b12d
Fix unresolved-attribute, not-subcriptable, invalid-type-form, possib…
elbersb Jan 13, 2026
287bd0b
Fix no-matching-overload, invalid-assignment, invalid-return-type
elbersb Jan 13, 2026
ce3ce3f
Fix invalid argument types
elbersb Jan 13, 2026
8be7cfb
Fix: invalid-return-type
elbersb Jan 13, 2026
9c7d194
Fix unsupported-operator errors
elbersb Jan 13, 2026
9753da4
Start tackling invalid-argument-type
elbersb Jan 13, 2026
3f11dda
Added more optional types
elbersb Jan 14, 2026
c1935ee
Fix remaining invalid-argument-type
elbersb Jan 14, 2026
30e0aef
First round of fixing tests
elbersb Jan 14, 2026
c35b393
Fix more test errors
elbersb Jan 14, 2026
51523ad
Fix remaining checks
elbersb Jan 14, 2026
d9fb120
Fixes required for py 3.9 specifically (old pandas type stubs)
elbersb Jan 14, 2026
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
16 changes: 11 additions & 5 deletions CLAUDE.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -16,6 +16,10 @@ uv pip install -e . --group dev

### Testing
```bash
# IMPORTANT Run all tests across Python versions
# to make sure all code changes work on older Python versions
uv run tox -p auto

# Run all tests with coverage
uv run pytest

Expand All @@ -27,9 +31,6 @@ uv run pytest tests/frequentist/test_z_test.py

# Run specific test
uv run pytest tests/frequentist/test_z_test.py::test_name

# Run all tests across Python versions
uv run tox
```

### Code Quality
Expand All @@ -40,8 +41,11 @@ uv run ruff check
# Run formatting
uv run ruff format

# Run type checking
uv run ty check

# Run all quality checks (as done in CI)
uv run ruff check && uv run ruff format && uv run pytest
uv run ruff check && uv run ruff format --check && uv run ty check && uv run pytest
```

### Build
Expand Down Expand Up @@ -148,4 +152,6 @@ The project uses `tox-uv` to leverage uv's fast package installation and environ

## Code Style

Uses ruff linting and formatting.
- **Linting & Formatting**: Uses [ruff](https://github.com/astral-sh/ruff) for code linting and formatting
- **Type Checking**: Uses [ty](https://github.com/astral-sh/ty) (Astral's fast Python type checker) for static type analysis
- Both tools are configured in `pyproject.toml` and exclude the `examples/` directory
1 change: 1 addition & 0 deletions CONTRIBUTING.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -91,6 +91,7 @@ Ready to contribute? Here's how to set up `confidence` for local development.

$ uv run ruff format # Format code
$ uv run ruff check # Lint code
$ uv run ty check # Type check code
$ uv run pytest # Run tests

To test across all supported Python versions (3.9, 3.10, 3.11, 3.12)::
Expand Down
7 changes: 7 additions & 0 deletions pyproject.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -35,6 +35,7 @@ dev = [
"pytest-cov>=4.0.0",
"pytest-xdist>=3.0.2",
"coverage>=7.0.0",
"ty>=0.0.11",
]

[project.urls]
Expand All @@ -53,6 +54,12 @@ extend-exclude = ["examples/"]
[tool.ruff.format]
quote-style = "double"

[tool.ty.src]
exclude = ["examples/"]

[tool.ty.rules]
unused-ignore-comment = "ignore"

[tool.pytest.ini_options]
addopts = "-v -n auto --cov=spotify_confidence --cov-report=html --cov-report=xml --cov-report=term-missing"
testpaths = ["tests"]
Expand Down
36 changes: 19 additions & 17 deletions spotify_confidence/analysis/abstract_base_classes/confidence_computer_abc.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -13,7 +13,7 @@
# limitations under the License.

from abc import ABC, abstractmethod
from typing import Iterable, List, Tuple, Union
from typing import Iterable, List, Optional, Tuple, Union

from pandas import DataFrame

Expand All @@ -32,11 +32,11 @@ def compute_difference(
level_1: Union[str, Iterable],
level_2: Union[str, Iterable],
absolute: bool,
groupby: Union[str, Iterable],
nims: NIM_TYPE,
final_expected_sample_size_column: str,
groupby: Optional[Union[str, Iterable]],
nims: Optional[NIM_TYPE],
final_expected_sample_size_column: Optional[str],
verbose: bool,
mde_column: str,
mde_column: Optional[str],
) -> DataFrame:
"""Return dataframe containing the difference in means between
group 1 and 2, p-value and confidence interval
Expand All @@ -46,42 +46,44 @@ def compute_difference(
@abstractmethod
def compute_multiple_difference(
self,
level: Union[str, Iterable],
level: Union[str, Iterable, int],
absolute: bool,
groupby: Union[str, Iterable],
level_as_reference: bool,
nims: NIM_TYPE,
final_expected_sample_size_column: str,
groupby: Optional[Union[str, Iterable]],
level_as_reference: Optional[bool],
nims: Optional[NIM_TYPE],
final_expected_sample_size_column: Optional[str],
verbose: bool,
mde_column: str,
mde_column: Optional[str],
) -> DataFrame:
"""Return dataframe containing the difference in means between
level and all other groups, with p-value and confidence interval
"""
pass

@abstractmethod
def compute_differences(
self,
levels: List[Tuple],
levels: Union[Tuple, List[Tuple]],
absolute: bool,
groupby: Union[str, Iterable],
nims: NIM_TYPE,
final_expected_sample_size_column: str,
groupby: Optional[Union[str, Iterable]],
nims: Optional[NIM_TYPE],
final_expected_sample_size_column: Optional[str],
verbose: bool,
mde_column: str,
mde_column: Optional[str],
) -> DataFrame:
"""Return dataframe containing the difference in means between
level and all other groups, with p-value and confidence interval
"""
pass

@abstractmethod
def achieved_power(
self,
level_1: Union[str, Iterable],
level_2: Union[str, Iterable],
mde: float,
alpha: float,
groupby: Union[str, Iterable],
groupby: Optional[Union[str, Iterable]],
) -> DataFrame:
"""Calculated the achieved power of test of differences between
level 1 and level 2 given a targeted MDE.
Expand Down
22 changes: 11 additions & 11 deletions spotify_confidence/analysis/abstract_base_classes/confidence_grapher_abc.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -13,7 +13,7 @@
# limitations under the License.

from abc import ABC, abstractmethod
from typing import Iterable, Union
from typing import Iterable, Optional, Union

from pandas import DataFrame

Expand All @@ -29,13 +29,13 @@ def __init__(
data_frame: DataFrame,
numerator_column: str,
denominator_column: str,
categorical_group_columns: str,
ordinal_group_column: str,
categorical_group_columns: Union[str, Iterable],
ordinal_group_column: Optional[str],
):
pass

@abstractmethod
def plot_summary(self, summary_df: DataFrame, groupby: Union[str, Iterable]) -> ChartGrid:
def plot_summary(self, summary_df: DataFrame, groupby: Optional[Union[str, Iterable]]) -> ChartGrid:
"""Plot for each group in the data_frame:

if ordinal level exists:
Expand All @@ -57,8 +57,8 @@ def plot_difference(
self,
difference_df: DataFrame,
absolute: bool,
groupby: Union[str, Iterable],
nims: NIM_TYPE,
groupby: Optional[Union[str, Iterable]],
nims: Optional[NIM_TYPE],
use_adjusted_intervals: bool,
split_plot_by_groups: bool,
) -> ChartGrid:
Expand All @@ -79,8 +79,8 @@ def plot_differences(
self,
difference_df: DataFrame,
absolute: bool,
groupby: Union[str, Iterable],
nims: NIM_TYPE,
groupby: Optional[Union[str, Iterable]],
nims: Optional[NIM_TYPE],
use_adjusted_intervals: bool,
split_plot_by_groups: bool,
) -> ChartGrid:
Expand All @@ -101,9 +101,9 @@ def plot_multiple_difference(
self,
difference_df: DataFrame,
absolute: bool,
groupby: Union[str, Iterable],
level_as_reference: bool,
nims: NIM_TYPE,
groupby: Optional[Union[str, Iterable]],
level_as_reference: Optional[bool],
nims: Optional[NIM_TYPE],
use_adjusted_intervals: bool,
split_plot_by_groups: bool,
) -> ChartGrid:
Expand Down
96 changes: 6 additions & 90 deletions spotify_confidence/analysis/bayesian/bayesian_base.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -27,7 +27,7 @@

# warnings.simplefilter("once")

INITIAL_RANDOMIZATION_SEED = np.random.get_state()[1][0]
INITIAL_RANDOMIZATION_SEED = np.random.get_state()[1][0] # type: ignore[index]


def num_decimals(value: float, absolute: bool) -> int:
Expand Down Expand Up @@ -114,6 +114,11 @@ def __init__(
self._all_group_columns = [column for column in self._all_group_columns if column is not None]
self._validate_data()

@abstractmethod
def _interval(self, row):
"""Return confidence/credible interval for a row. Must be implemented by subclasses."""
pass

def _validate_data(self):
"""Integrity check input dataframe."""
if not self._all_group_columns:
Expand All @@ -137,36 +142,6 @@ def _validate_data(self):
Must be number or datetime type.""".format(ordinal_column_type)
)

@classmethod
def as_cumulative(
cls, data_frame, numerator_column, denominator_column, ordinal_group_column, categorical_group_columns=None
):
"""
Instantiate the class with a cumulative representation of the dataframe.
Sorts by the ordinal variable and calculates the cumulative sum
May be used for to visualize the difference between groups as a
time series.

Args:
data_frame (pd.DataFrame): DataFrame
numerator_column (str): Column name for numerator column.
denominator_column (str): Column name for denominator column.
ordinal_group_column (str): Column name for ordinal grouping
(e.g. numeric or date values).
categorical_group_columns (str or list),
Optional: Column names for categorical groupings.

"""

sorted_df = data_frame.sort_values(by=ordinal_group_column)
cumsum_cols = [numerator_column, denominator_column]
if categorical_group_columns:
sorted_df[cumsum_cols] = sorted_df.groupby(by=categorical_group_columns)[cumsum_cols].cumsum()
else:
sorted_df[cumsum_cols] = sorted_df[cumsum_cols].cumsum()

return cls(sorted_df, numerator_column, denominator_column, categorical_group_columns, ordinal_group_column)

def summary(self):
"""Return Pandas DataFrame with summary statistics."""
return self._summary(self._data_frame, self._interval)
Expand Down Expand Up @@ -490,13 +465,6 @@ def _iterate_groupby_to_dataframe(self, input_function, groupby, **kwargs):

return results_data_frame

def _all_groups(self):
"""Return a list of all group keys.

Returns: list"""
groups = list(self._data_frame.groupby(self._all_group_columns).groups.keys())
return groups

def _add_group_by_columns(self, difference_df, groupby, level_name):
if groupby:
groupby = groupby[0] if len(groupby) == 1 else groupby
Expand All @@ -505,55 +473,3 @@ def _add_group_by_columns(self, difference_df, groupby, level_name):
else:
for col, val in zip(groupby, level_name):
difference_df.insert(0, column=col, value=val)


# class BinomialResponse(BaseTest, metaclass=ABCMeta):
# """Binomial Response Variable.
# """

# class GaussianResponse(BaseTest, metaclass=ABCMeta):
# """Base class for tests of normal response variables

# E.g. Revenue per user
# """

# pass


# class PoissonResponse(BaseTest, metaclass=ABCMeta):
# """Base class for tests of poisson response variables.

# E.g. # of days active per user per month
# """
# pass


# class MultinomialResponse(BaseTest, metaclass=ABCMeta):
# """Base class for tests of multinomial response variables.

# E.g. single choice answer survey
# self.
# """

# def __init__(self, data_frame, categorical_group_columns,
# ordinal_group_column, category_column, value_column):
# self._category_column = category_column
# self._value_column = value_column
# super().__init__(data_frame, categorical_group_columns,
# ordinal_group_column)


# class CategoricalResponse(BaseTest, metaclass=ABCMeta):
# """Base class for tests of categorical response variables.

# E.g. multiple choice answer survey
# """

# def __init__(self, data_frame, categorical_group_columns,
# ordinal_group_column, category_column, value_column):
# self._category_column = category_column
# self._value_column = value_column
# super().__init__(data_frame, categorical_group_columns,
# ordinal_group_column)

# pass
35 changes: 0 additions & 35 deletions spotify_confidence/analysis/bayesian/bayesian_models.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -735,38 +735,3 @@ def _categorical_multiple_difference_plot(self, level, absolute, groupby, level_
)

return results_df


# class GammaPoisson(PoissonResponse):
# pass


# class DirichetMultinomial(MultinomialResponse):
# def __init__(self,
# data_frame,
# group_columns,
# category_column,
# value_column,
# prior_value_column=None):

# super().__init__(data_frame, group_columns, category_column,
# value_column)


# class Gaussian(GaussianResponse):
# def __init__(self,
# data_frame,
# groupings,
# mean_col,
# std_col,
# n_col,
# time_grouping=None,
# prior_columns=None):
# self.prior_lambda_column = prior_lambda_column
# super(BaseGaussianResponse, self).__init__(
# data_frame, groups, mean_col, std_col, n_col, time_grouping)
# raise (NotImplementedError)


# class DirichetCategorical(CategoricalResponse):
# pass
Loading