data-science-types

0.2.23 verified Fri May 01 auth: no python maintenance

Type stubs for Python machine learning libraries such as pandas, numpy, matplotlib, and scikit-learn. Current version 0.2.23, requires Python >=3.6. Low release cadence.

pip install data-science-types

Common errors

error Cannot find implementation or library stub for module 'pandas' ↓

cause Mypy cannot locate stub files; data-science-types may not be installed or is not being recognized.

fix

Run pip install data-science-types and ensure your mypy configuration includes packages.

error Missing type hints for pandas.DataFrame.groupby ↓

cause Stub coverage incomplete; data-science-types does not define all methods.

fix

Use pandas-stubs for more complete coverage, or add inline type hints with # type: ignore.

Warnings

gotcha The package does not provide stubs for all functions and may be incomplete. It is not actively maintained; consider migrating to typeshed stubs (pandas-stubs, numpy-stubs) instead. ↓

fix Install individual stub packages: pip install pandas-stubs numpy-stubs

deprecated data-science-types is in maintenance mode with infrequent updates. Newer pandas/numpy versions may have incomplete stub coverage. ↓

fix Use alternative stub packages like pandas-stubs, numpy-stubs, and scikit-learn-stubs.

Imports

numpy
```
import numpy as np
```
Type stubs are automatically used by mypy/pyright when installed, no special import needed.
pandas
```
import pandas as pd
```
Same as numpy; stubs are activated by package presence.

Quickstart

Basic usage: install package, then use pandas/numpy as usual; mypy or pyright will use stubs.

import pandas as pd
import numpy as np

df = pd.DataFrame({'a': [1, 2, 3]})
reveal_type(df)  # mypy will infer DataFrame type