OpenBB-finance/OpenBB · error · TA_DataException

Error processing indicator {indicator.name}: {e}

Error message

Error processing indicator {indicator.name}: {e}

What it means

In PlotlyTA.get_indicator_output-ish aggregation (data_classes.py:386), each indicator's data is produced by self.get_indicator_data(...); any exception it raises is wrapped into TA_DataException(f"Error processing indicator {indicator.name}: {e}") with the original chained. The inner error is usually a missing input column (high/low/volume) for that specific indicator, NaN-laden data, or a pandas_ta computation failure.

Source

Thrown at openbb_platform/obbject_extensions/charting/openbb_charting/core/plotly_ta/data_classes.py:386

        output = self.df_ta
        for indicator in active_indicators:
            if (
                indicator.name in self.columns
                and "volume" in self.columns[indicator.name]
                and not self.has_volume
            ):
                continue
            if indicator.name in ["fib", "srlines", "clenow", "demark", "ichimoku"]:
                continue
            try:
                indicator_data = self.get_indicator_data(
                    indicator,
                    **self.indicators.get_options_dict(indicator.name) or {},
                )
            except Exception as e:
                indicator_data = None
                raise TA_DataException(
                    f"Error processing indicator {indicator.name}: {e}"
                ) from e

            if indicator_data is not None:
                output = output.join(indicator_data).infer_objects()
                numeric_cols = output.select_dtypes(include=["number"]).columns
                output[numeric_cols] = output[numeric_cols].interpolate("linear")

        return output

View on GitHub (pinned to 3e071fcc2c)

Solutions

1. Inspect the __cause__ of the TA_DataException — it names the indicator and the underlying failure.
2. Ensure the frame has the columns the indicator needs (high, low, close, volume) or drop that indicator from the request.
3. Clean NaNs / ensure enough rows for the indicator's lookback window.
4. Catch TA_DataException per run and fall back to a reduced indicator set so one bad indicator doesn't kill the chart.

Example fix

# before
indicators = {"obv": {}, "sma": {"length": 20}}  # fails on close-only data
fig = obbject.charting.to_chart(data=df, indicators=indicators)

# after
indicators = {"sma": {"length": 20}}  # drop volume-dependent obv
fig = obbject.charting.to_chart(data=df, indicators=indicators)

required = {"obv": ["close", "volume"], "atr": ["high", "low", "close"], "adx": ["high", "low", "close"]}
have = set(map(str.lower, df.columns))
indicators = {k: v for k, v in indicators.items() if set(required.get(k, ["close"])) <= have}

def indicator_supported(cols: set[str], name: str) -> bool:
    needs = {"obv": {"close", "volume"}, "ad": {"high", "low", "close", "volume"}}.get(name, {"close"})
    return needs <= cols

from openbb_charting.core.plotly_ta.data_classes import TA_DataException
try:
    output = ta.get_base_data()  # or full aggregation
except TA_DataException as e:
    logger.warning("indicator failed (%s), cause=%s", e, e.__cause__)
    indicators.pop(bad_name, None)  # retry with reduced set

When it happens

Trigger: Requesting an indicator whose required columns are absent — e.g. 'obv'/'ad' without a volume column, 'atr' without high/low — or indicator option dicts from ChartIndicators containing invalid values for pandas_ta.

Common situations: Charting index or economic data (close-only series) with a default indicator set that assumes OHLCV; sparse frames where rolling windows produce all-NaN; stale indicator option names after a pandas_ta upgrade.

Related errors

• No close column found in dataframe
• Unknown indicator: {indicator}
• Please make sure that the columns 'High', 'Low', and 'Close
• Failed to convert results to chart. Ensure the provided data
• Please provide data with only one symbol and columns for OHL

AI-assisted analysis of OpenBB-finance/OpenBB@3e071fcc2c (2026-08-14). Data as JSON: /api/errors/143663d0a3cb8d5f. Report an issue: GitHub.