run_feature_worker

Overview

run_feature_worker computes technical indicators (TA-Lib) from OHLCV data. It is part of the optional ML path — not required for equal weight / HRP / DCC backtests that go data → PO → trading. Ask for the indicators you care about (e.g. “RSI and MACD only”); skills build the full Features params config.

Parameters

string

required

Blob storage URL pointing to a data_extractor_*.json file.This is the output_url returned by run_data_extraction.

object

required

Feature engineering configuration.

Show config fields

object

required

Controls which technical indicators are computed. Each indicator group is an object with a status boolean and a windows array of integer periods.

Show Indicator groups

object

Bollinger Bands (upper, middle, lower). Fields: status (bool), windows (int array). Columns produced: bb_upper_{w}, bb_middle_{w}, bb_lower_{w}.

object

Simple Moving Average (SMA) and Exponential Moving Average (EMA). Fields: status (bool), windows (int array). Columns: sma_{w}, ema_{w}.

object

Relative Strength Index (RSI). Fields: status (bool), windows (int array). Columns: rsi_{w}.

object

MACD line, signal, and histogram. fastperiod = w, slowperiod = 2*w, signalperiod = 9. Fields: status (bool), windows (int array). Columns: macd_{w}, macd_signal_{w}, macd_hist_{w}.

object

Stochastic Oscillator (Stoch K and D). Fields: status (bool), windows (int array). Columns: slowk_{w}, slowd_{w} (TA-Lib STOCH).

object

Average True Range (ATR). Fields: status (bool), windows (int array). Columns: atr_{w}.

object

Average Directional Index (ADX). Fields: status (bool), windows (int array). Columns: adx_{w}.

object

Commodity Channel Index (CCI). Fields: status (bool), windows (int array). Columns: cci_{w}.

object

Rate of Change (ROC). Fields: status (bool), windows (int array). Columns: roc_{w}.

boolean

On-Balance Volume (OBV). Set to true to enable. Produces a single obv column.

object

Candlestick pattern recognition (TA-Lib CDL* functions). Set status: true to enable, then individually toggle each pattern name (e.g. "Doji": true, "Hammer": true). Over 50 patterns are available.

Returns

{
  "status": "Succeeded",
  "output_url": "https://stqsnpprod.blob.core.windows.net/data/feature_engine_a1c3e5f70912.json",
  "output_name": "feature_engine_a1c3e5f70912.json",
  "execution_name": "job-feature-worker-abc123xyz"
}

Field	Description
`status`	Job terminal status (`Succeeded`)
`output_url`	Full HTTPS URL to the output blob — pass to `run_ml_job` or `run_dl_job`
`output_name`	Blob filename
`execution_name`	Job execution ID for audit/debugging

Example

{
  "input_url": "https://stqsnpprod.blob.core.windows.net/data/data_extractor_2b510101b9b7.json",
  "config": {
    "Features params": {
      "bb_lines": { "status": false, "windows": [20] },
      "moving_averages": { "status": true, "windows": [5, 15, 22] },
      "rsi_indicators": { "status": true, "windows": [5, 15] },
      "macd_indicators": { "status": true, "windows": [12] },
      "stochastic_indicators": { "status": false, "windows": [14] },
      "atr_indicators": { "status": false, "windows": [14] },
      "adx_indicators": { "status": false, "windows": [14] },
      "cci_indicators": { "status": false, "windows": [14] },
      "roc_indicators": { "status": false, "windows": [10] },
      "obv_indicator": false,
      "candlestick_patterns": { "status": false }
    }
  }
}

Resources

Resource	Value
Container Apps Job	`job-feature-worker`
Container name	`feature-worker`
Env vars injected	`INPUT_URL`, `CONFIG`
Output blob prefix	`feature_engine_`

Next Step

Pass output_url (with the original data_extractor_ URL) to:

run_ml_job — sklearn / XGBoost / LightGBM / CatBoost regressors
run_dl_job — PyTorch networks

Overview

How To

Tools description

Overview

Parameters

Returns

Example

Resources

Next Step

​Overview

​Parameters

​Returns

​Example

​Resources

​Next Step

Overview

Parameters

Returns

Example

Resources

Next Step