polars_talib

repository·master·Indexed 42 days ago

https://github.com/yvictor/polars_ta_extension

A high-performance Polars extension for the TA-Lib technical analysis library, providing fast, vectorized technical indicator calculations within Polars DataFrames. Version 0.1.6 supports a wide range of indicators across Overlap Studies, Momentum, Volume, Cycle, Price Transform, Volatility, Pattern Recognition, and Statistic functions. It integrates directly into Polars expressions via a .ta namespace or standalone API calls, and supports grouped calculations using the .over() syntax.

Tokens: 16.1K
Snippets: 72
Records: 94
Agent score: 53%

What's inside polars_talib

Import polars_talib
master
To use the extension, import polars and polars_talib (commonly aliased as plta).
```
import polars
import polars_talib as plta
```
https://github.com/yvictor/polars_ta_extension/blob/master/README.md
Install the talib Rust crate
master
To use the safe Rust wrapper for Ta-Lib in your project, add the talib crate using cargo.
```
cargo add talib
```
https://github.com/yvictor/polars_ta_extension/blob/master/talib/README.md

Use talib in Rust

master

To use talib, you must follow a lifecycle of initializing the library, performing calculations using specific indicator functions, and then shutting down the library.

Key steps:

Initialize: Call ta_initialize() at the start of your application.
Configure: Use TimePeriodKwargs or TimePeriodKwargsBuilder to specify parameters like timeperiod.
Calculate: Pass a pointer to your data (e.g., Vec<f64>), the length of the data, and the configuration object to the indicator function (e.g., ta_rsi).
Shutdown: Call ta_shutdown() when the application is finished with Ta-Lib.

use talib::common::{TimePeriodKwargs, ta_initialize, ta_shutdown};
use talib::momentum::ta_rsi;

fn main() {
    // Initialize Ta-Lib
    let _ = ta_initialize();

    // Sample close prices
    let close_prices: Vec<f64> = vec![1.087010, 1.087120, 1.087080, 1.087170, 1.087110, 1.087010, 1.087100, 1.087120, 1.087110, 1.087080, 1.087000, 1.086630, 1.086630, 1.086610, 1.086630, 1.086640, 1.086650, 1.086650, 1.086670, 1.086630];

    // Specify the RSI calculation parameters using a builder
    let kwargs = TimePeriodKwargsBuilder::default()
        .timeperiod(5)
        .build()
        .unwrap();

    // Calculate RSI using a pointer to the data
    let res = ta_rsi(close_prices.as_ptr(), close_prices.len(), &kwargs);

    // Process the result
    match res {
        Ok(rsi_values) => {
            for (index, value) in rsi_values.iter().enumerate() {
                println!("RSI at index {}: {}", index, value);
            }
        }
        Err(e) => {
            println!("Error: {:?}", e);
        }
    }

    // Shutdown Ta-Lib
    let _ = ta_shutdown();
}

Install polars_talib
master
Install the extension using pip:
```
pip install polars_talib
```
https://github.com/yvictor/polars_ta_extension/blob/master/README.md
Momentum indicators returning a Struct
master
Certain momentum indicators return multiple values encapsulated in a Polars Struct. You can access the individual components by name.
Aroon
Returns a struct with fields:
- aroondown: Float64
- aroonup: Float64
MACD (Moving Average Convergence Divergence)
Returns a struct with fields:
- macd: Float64
- macdsignal: Float64
- macdhist: Float64
- Note: macdext and macdfix variants also return this same structure.
Stochastic (stoch)
Returns a struct with fields:
- slowk: Float64
- slowd: Float64
Stochastic Fast (stochf)
Returns a struct with fields:
- fastk: Float64
- fastd: Float64
Stochastic RSI (stochrsi)
Returns a struct with fields:
- fastk: Float64
- fastd: Float64
https://github.com/yvictor/polars_ta_extension/blob/master/src/momentum.rs
Use Momentum indicators in Polars
master
The polars_ta_extension provides several momentum indicators as Polars expressions. These indicators can be used within Polars lazy or eager API calls to compute technical analysis metrics on financial time series data.
Most momentum indicators require a single input series (e.g., close), while others require multiple series (e.g., high, low, close, volume). Some indicators return a single Float64 series, while others return a Struct containing multiple components.
https://github.com/yvictor/polars_ta_extension/blob/master/src/momentum.rs
Use polars_talib for technical analysis with Polars
master
The polars_talib extension allows you to perform TA-Lib technical analysis directly within Polars expressions. A key pattern for handling multiple symbols (e.g., stock tickers) in a single DataFrame is to use the .over("Symbol") syntax. This ensures that indicators like SMA or MACD are calculated independently for each group.
To use default column names (like open, high, low, close) without explicit mapping, it is recommended to lowercase your float columns using Polars selectors.
```
import polars as pl
import polars_talib as plta

# Prepare data by lowercasing float columns to match TA-Lib defaults
p = pl.scan_parquet("data.parquet").select(
    pl.col("Date"), 
    pl.col("Ticker").alias("Symbol"),
    pl.selectors.float().name.to_lowercase()
)

# Apply indicators using the .over() syntax for grouped calculations
df = p.with_columns(
    plta.sma(timeperiod=5).over("Symbol").alias("sma5"),
    plta.macd(fastperiod=10, slowperiod=20, signalperiod=5).over("Symbol").alias("macd"),
).collect()
```
https://github.com/yvictor/polars_ta_extension/blob/master/examples/basic.ipynb
Use technical indicators with Polars expressions
master
You can use technical indicators directly within Polars expressions using the .ta namespace on a column. This allows for seamless integration into with_columns calls.
Note that some indicators like macd return a struct containing multiple fields (e.g., macd, macdsignal, macdhist), which can be accessed using .struct.field().
```
df.with_columns(
    pl.col("close").ta.ema(5).alias("ema5"),
    pl.col("close").ta.macd(12, 26, 9).struct.field("macd"),
    pl.col("close").ta.macd(12, 26, 9).struct.field("macdsignal"),
    pl.col("open").ta.cdl2crows(pl.col("high"), pl.col("low"), pl.col("close")).alias("cdl2crows"),
    pl.col("close").ta.wclprice("high", "low").alias("wclprice"),
)
```
https://github.com/yvictor/polars_ta_extension/blob/master/README.md

Calculate indicators for multiple symbols using .over()

master

When working with datasets containing multiple symbols (e.g., a 'symbol' or 'Ticker' column), use the Polars .over() syntax to ensure indicators are calculated independently for each group.

df.with_columns(
    pl.col("close").ta.ema(5).over("symbol").alias("ema5"),
    pl.col("close").ta.macd(12, 26, 9).over("symbol").struct.field("macd"),
    pl.col("close").ta.macd(12, 26, 9).over("symbol").struct.field("macdsignal"),
    pl.col("open").ta.cdl2crows(
        pl.col("high"), pl.col("low"), pl.col("close")
    ).over("symbol").alias("cdl2crows"),
    pl.col("close").ta.wclprice("high", "low").over("symbol").alias("wclprice"),
)

Use polars_talib functions as standalone API calls
master
In addition to the .ta namespace, you can use polars_talib (as plta) functions directly. This provides a flexible interface similar to talib.abstract. You can pass specific columns or allow the function to use default columns if supported.
```
df.with_columns(
    plta.ht_dcperiod(),
    plta.ht_dcperiod(pl.col("close")),
    plta.aroon(),
    plta.aroon(pl.col("high"), pl.col("low"), timeperiod=10),
    plta.wclprice(),
    plta.wclprice(
        pl.col("high"), pl.col("low"), pl.col("close"), 
        timeperiod=10
    ),
)
```
https://github.com/yvictor/polars_ta_extension/blob/master/README.md
List indicator functions grouped by category
master
To see which functions belong to which indicator groups (e.g., Momentum, Volatility, Pattern Recognition), use plta.get_function_groups().
```
import polars_talib as plta

# dict of functions by group
plta.get_function_groups()
```
https://github.com/yvictor/polars_ta_extension/blob/master/README.md
List supported indicator functions
master
You can retrieve a list of all available technical analysis functions using plta.get_functions().
```
import polars_talib as plta

# list of functions
plta.get_functions()
```
https://github.com/yvictor/polars_ta_extension/blob/master/README.md

polars_talib

What's inside polars_talib

Import polars_talib

Install the talib Rust crate

Use talib in Rust

Install polars_talib

Momentum indicators returning a Struct

Aroon

MACD (Moving Average Convergence Divergence)

Stochastic (`stoch`)

Stochastic Fast (`stochf`)

Stochastic RSI (`stochrsi`)

Use Momentum indicators in Polars

Use polars_talib for technical analysis with Polars

Use technical indicators with Polars expressions

Calculate indicators for multiple symbols using .over()

Use polars_talib functions as standalone API calls

List indicator functions grouped by category

List supported indicator functions

polars_talib

What's inside polars_talib

Aroon

MACD (Moving Average Convergence Divergence)

Stochastic (stoch)

Stochastic Fast (stochf)

Stochastic RSI (stochrsi)

Stochastic (`stoch`)

Stochastic Fast (`stochf`)

Stochastic RSI (`stochrsi`)