Docs

BDS+Polars

I already have raw cycler files. How do I prepare them for analysis in Polars?

BDS+Polars

Question

I already have raw cycler files. How do I prepare them for analysis in Polars?

Use This When

Use this path when you want fast dataframe operations in Python and you want BDS to normalize raw cycler exports before analysis.

File-Based Path

For reusable analysis, convert to Parquet:

bds convert raw.mpt normalized.polars.parquet --target polars --format parquet --current-sign preserve

Then read lazily with Polars:

import polars as pl
 
lf = pl.scan_parquet("normalized_exports/**/*.polars.parquet")
 
cycle_summary = (
    lf.group_by("Cycle Count")
    .agg(
        pl.max("Voltage (V)").alias("max_voltage_v"),
        pl.min("Voltage (V)").alias("min_voltage_v"),
        pl.max("Discharging Capacity (Ah)").alias("discharge_capacity_ah"),
    )
    .collect()
)

In-Memory Path

If you already have an in-memory dataframe from bds.read(), convert it to the same BDS export shape before writing Polars expressions intended for files:

import battery_data_standard as bds
import polars as pl
from battery_data_standard.export import to_export_frame
 
df = bds.read("raw.mpt", cycler="auto", current_sign="preserve")
bds_df = to_export_frame(df, target="polars")
 
cycle_summary = bds_df.group_by("Cycle Count").agg(
    pl.max("Voltage (V)").alias("max_voltage_v"),
    pl.min("Voltage (V)").alias("min_voltage_v"),
    pl.max("Discharging Capacity (Ah)").alias("discharge_capacity_ah"),
)

EIS Data

bds convert-eis impedance.csv normalized.eis.parquet --format parquet

import polars as pl
 
eis = pl.read_parquet("normalized.eis.parquet")

The standardized EIS columns are Frequency_Hz, Zre_exp_Ohm, and Zim_exp_Ohm.

Known Limits

Prefer bds convert --target polars --format parquet for shareable files.
If you use bds.read() in memory, call to_export_frame(..., target="polars") before copying file-based examples.