reflex-plot

Seamlessly plot dataframes in Reflex just like you would do with Matplotlib.

You can import plot from reflex_plot to get a Rechart component.

import random
from typing import Literal

import pandas as pd
import reflex as rx
from reflex_plot import plot

def plot_data(kind: Literal["line", "area", "bar"]) -> rx.Component:
    df = pd.DataFrame(
        {
            "category": list(range(20)),
            "value": [random.randint(0, 1000) for _ in range(20)],
        }
    )
    return plot(
        df,
        kind=kind,
        x="category",
        y="value",
        grid=True,
        tool_tip=True,
    )

Plot any DataFrame

Using Narwhals under the hood, reflex-plot supports plotting from any DataFrame library including Pandas, Polars, and Dask. The plot() function accepts any DataFrame type and seamlessly converts it to a Recharts component.

import polars as pl
import pandas as pd
import dask.dataframe as dd

# These all work the same way
df_pd = pd.DataFrame({"x": [1,2,3], "y": [4,5,6]})
df_pl = pl.DataFrame({"x": [1,2,3], "y": [4,5,6]})
df_dd = dd.from_pandas(df_pd)

plot(df_pd, kind="line", x="x", y="y")
plot(df_pl, kind="line", x="x", y="y")
plot(df_dd, kind="line", x="x", y="y")

Pandas plotting backend

You can also set reflex_plot as the default backend for pandas and use DataFrame.plot as you would do with Matplotlib

pd.set_option("plotting.backend", "reflex_plot")


def plot_data(kind: Literal["line", "area", "bar"]) -> rx.Component:
    df = pd.DataFrame(
        {
            "category": list(range(20)),
            "value": [random.randint(0, 1000) for _ in range(20)],
        }
    )
    return df.plot(
        kind=kind,
        x="category",
        y="value",
        grid=True,
        tool_tip=True,
    )

How to install

pip install reflex-plot

Charts type covered

• Line
• Area
• Bar