Chart widget¶

Browse Yahoo Finance prices with centered Ticker and Max bars controls. The widget handles the input layout and chart updates; the notebook supplies a price loader.

Install the optional notebook dependencies and Yahoo client in your notebook environment:

pip install 'mplchart[notebook,pandas]' yfinance

Run this notebook with a live kernel and widget support to use the controls. Fetching Yahoo data requires internet access; a static preview cannot run the loader.

Load Yahoo prices¶

chart_widget accepts any callable that takes a ticker and returns a pandas or Polars prices DataFrame. Here yfinance supplies pandas data, and normalize_prices prepares its columns and date index for mplchart.

Fetch five years so moving averages have history before the visible window. The small in-memory cache avoids another Yahoo request when revisiting a ticker. Run get_prices.cache_clear() before creating a new widget to refresh cached prices.

In [3]:

from functools import lru_cache

import pandas as pd
import yfinance as yf

from mplchart.notebook import chart_widget
from mplchart.utils import normalize_prices


@lru_cache(maxsize=32)
def get_prices(ticker: str, period: str = "5y") -> pd.DataFrame:
    prices = yf.Ticker(ticker).history(period=period, auto_adjust=True)
    if prices.empty:
        raise ValueError(f"No Yahoo prices found for {ticker!r}")
    return normalize_prices(prices)

from functools import lru_cache import pandas as pd import yfinance as yf from mplchart.notebook import chart_widget from mplchart.utils import normalize_prices @lru_cache(maxsize=32) def get_prices(ticker: str, period: str = "5y") -> pd.DataFrame: prices = yf.Ticker(ticker).history(period=period, auto_adjust=True) if prices.empty: raise ValueError(f"No Yahoo prices found for {ticker!r}") return normalize_prices(prices)

Browse a chart¶

Enter a ticker such as AAPL, MSFT, or SPY, then press Enter or leave the field to update the chart. Max bars controls only the visible window and reuses the loaded history.

The simplest call is chart_widget(get_prices), which shows candlesticks and volume. This example supplies a full plot sequence with two moving averages and an RSI pane. Leave the widget as the last expression in the cell to display it.

In [4]:

from mplchart.indicators import SMA, RSI
from mplchart.primitives import Candlesticks, Volume, Pane, Line

indicators = [
    Candlesticks(),
    SMA(50),
    SMA(200),
    Volume(),
    Pane("below", yticks=(30, 50, 70)),
    Line(RSI(14), overbought=70, oversold=30),
]

chart_widget(get_prices, ticker="AAPL", max_bars=250, indicators=indicators)

from mplchart.indicators import SMA, RSI from mplchart.primitives import Candlesticks, Volume, Pane, Line indicators = [ Candlesticks(), SMA(50), SMA(200), Volume(), Pane("below", yticks=(30, 50, 70)), Line(RSI(14), overbought=70, oversold=30), ] chart_widget(get_prices, ticker="AAPL", max_bars=250, indicators=indicators)

Out[4]:

VBox(children=(HBox(children=(Text(value='AAPL', continuous_update=False, description='Ticker:'), IntText(valu…

Use another loader¶

For an existing bardata feed, pass feed.get instead of get_prices. To bind query options, use functools.partial, for example partial(feed.get, freq="weekly"). Use indicators from mplchart.expressions if your loader returns Polars data.

See the notebook API reference for chart options and loader behavior.