pola-rs/polars · error · TypeError

cannot convert DataFrame to

Error message

cannot convert DataFrame to {target} (mixed type columns result in `object` dtype)
{df.schema!r}

What it means

After conversion, if the NumPy array's dtype is `object`, the DataFrame had mixed-type columns that NumPy can only represent as object arrays, which is unsuitable for Jax/PyTorch. Polars raises instead of silently returning an object array.

Solutions

Select only numeric columns: df.select(pl.col(pl.Float64), ...) or df.select(pl.numeric())
Cast columns to a common dtype before conversion
Drop non-numeric columns like strings/datetimes

Example fix

// before
arr = df.to_torch()  # df has a 'name' string column
// after
arr = df.select(pl.numeric()).to_torch()

Defensive patterns

Strategy: validation

Validate before calling

import numpy as np
obj_cols = [n for n, t in df.schema.items() if not t.is_numeric()]
if obj_cols:
    df = df.select(pl.numeric())
arr = df.to_torch()

Type guard

def is_numeric_frame(df) -> bool:
    return all(tp.is_numeric() for tp in df.schema.dtypes())

Try / catch

try:
    arr = df.to_jax()
except TypeError:
    arr = df.select(pl.numeric()).to_jax()

Prevention

Select only numeric columns for tensor conversion
Drop string/ID/categorical columns before conversion
Cast mixed numeric dtypes to a common dtype (e.g. Float32)

When it happens

Trigger: Calling df.to_jax()/df.to_torch() on a DataFrame mixing incompatible dtypes (e.g. strings with numbers, or multiple different dtypes that produce object arrays).

Common situations: Feature frames that accidentally include an ID/string column alongside numeric features; datetime + float mixes.

Related errors

AI-assisted analysis of pola-rs/polars@fe841f959e (2026-09-18). Data as JSON: /api/errors/f5d84583bd4c78c1. Report an issue: GitHub.

Appendix: source

Thrown at py-polars/src/polars/ml/utilities.py:29

    *,
    writable: bool,
    target: str,
    order: IndexOrder = "fortran",
) -> np.ndarray[Any, Any]:
    """Convert a DataFrame to a NumPy array for use with Jax or PyTorch."""
    for nm, tp in df.schema.items():
        if tp == List:
            msg = f"cannot convert List column {nm!r} to {target} (use Array dtype instead)"
            raise TypeError(msg) from None

    if df.width == 1 and df.schema.dtypes()[0] == Array:
        arr = df[df.columns[0]].to_numpy(writable=writable)
    else:
        arr = df.to_numpy(writable=writable, order=order)

    if arr.dtype == object:
        msg = f"cannot convert DataFrame to {target} (mixed type columns result in `object` dtype)\n{df.schema!r}"
        raise TypeError(msg)
    return arr

View on GitHub (pinned to fe841f959e)