OpenBB-finance/OpenBB · error · ValueError

No close column found in dataframe

Error message

No close column found in dataframe

What it means

PlotlyTA(data_classes.py:276) requires an identifiable close column: it calls check_columns(df_ta) (plotly_ta/ta_helpers.py) which searches for a column matching 'Adj Close', 'adj_close', or 'Close' case-insensitively. If none is found the constructor raises ValueError("No close column found in dataframe") because nearly every technical indicator (ad, adosc, adx, atr, cci, kc, obv, ...) needs close prices.

Solutions

  1. Rename the price column before charting: df = df.rename(columns={"last": "close"}).
  2. If columns are MultiIndex/tuples, flatten them to plain strings first (df.columns = df.columns.get_level_values(0)).
  3. Pass a standard OHLC frame (open/high/low/close, volume) from the router instead of a hand-built one.
  4. Disable indicators if you only want candles: indicators={}

Example fix

# before
PlotlyTA(df_ta=df, indicators={"sma": {}})  # ValueError: No close column found in dataframe

# after
df = df.rename(columns={"last": "close"})
PlotlyTA(df_ta=df, indicators={"sma": {}})
Defensive patterns

Strategy: validation

Validate before calling

import re
CLOSE_RE = re.compile(r"(Adj\sClose|adj_close|Close)", re.IGNORECASE)

def has_close_column(df) -> bool:
    return any(CLOSE_RE.search(c) for c in map(str, df.columns))

if not has_close_column(df):
    df = df.rename(columns={"last": "close"})

Type guard

def ta_frame(df: pd.DataFrame) -> bool:
    cols = {str(c).lower() for c in df.columns}
    return bool(cols & {"close", "adj close", "adj_close"})

Try / catch

try:
    ta = PlotlyTA(df_ta=df, indicators=indicators)
except ValueError as e:
    if "close column" in str(e):
        raise ValueError("charting requires a close column; rename your price column") from e
    raise

Prevention

When it happens

Trigger: Constructing PlotlyTA (directly or via charting with indicators) on a DataFrame whose price column is named differently — e.g. 'last', 'price', '4. close', or a localized name — or one that only has volume.

Common situations: Feeding charting output from a custom provider whose Close field was renamed during transform_data; using intraday frames where the provider emits 'close' only after normalization that was skipped; passing symbol-keyed multi-level columns so str(columns) no longer matches the regex cleanly.

Related errors

AI-assisted analysis of OpenBB-finance/OpenBB@3e071fcc2c (2026-08-14). Data as JSON: /api/errors/c2be5f04c967c42b. Report an issue: GitHub.

Appendix: source

Thrown at openbb_platform/obbject_extensions/charting/openbb_charting/core/plotly_ta/data_classes.py:276

        ma_mode: list[str] | None = None,
    ):
        """Initialize."""
        # pylint: disable=import-outside-toplevel
        from pandas import DataFrame, Series  # noqa
        from openbb_charting.core.plotly_ta.ta_helpers import check_columns  # noqa

        if isinstance(df_ta, Series):
            df_ta = df_ta.to_frame()

        if not isinstance(indicators, ChartIndicators):
            indicators = ChartIndicators.from_dict(indicators)

        self.df_ta: DataFrame = df_ta
        self.indicators: ChartIndicators = indicators
        self.ma_mode: list[str] = ma_mode or ["sma", "ema", "wma", "hma", "zlma", "rma"]
        self.close_col = check_columns(df_ta)
        if self.close_col is None:
            raise ValueError("No close column found in dataframe")

        self.columns: dict[str, list[str]] = {
            "ad": ["high", "low", self.close_col, "volume"],
            "adosc": ["high", "low", self.close_col, "volume"],
            "adx": ["high", "low", self.close_col],
            "aroon": ["high", "low"],
            "atr": ["high", "low", self.close_col],
            "cci": ["high", "low", self.close_col],
            "donchian": ["high", "low"],
            "fisher": ["high", "low"],
            "kc": ["high", "low", self.close_col],
            "obv": [self.close_col, "volume"],
            "stoch": ["high", "low", self.close_col],
            "vwap": ["high", "low", self.close_col, "volume"],
        }

        self.has_volume = "volume" in df_ta.columns and bool(df_ta["volume"].sum() > 0)

View on GitHub (pinned to 3e071fcc2c)