mirror of
https://github.com/mihakralj/QuanTAlib.git
synced 2026-08-17 01:58:06 +00:00
416 lines
13 KiB
C#
416 lines
13 KiB
C#
using Xunit;
|
|
using System;
|
|
using QuanTAlib;
|
|
using System.Runtime.InteropServices;
|
|
using System.Runtime.InteropServices.Marshalling;
|
|
using Python.Runtime;
|
|
|
|
namespace Validations;
|
|
|
|
public class PandasTA : IDisposable {
|
|
private bool disposed = false;
|
|
private readonly GBM_Feed bars;
|
|
private readonly Random rnd = new();
|
|
private readonly int period, skip;
|
|
private readonly int digits;
|
|
private readonly dynamic np;
|
|
private readonly dynamic ta;
|
|
private readonly dynamic pd;
|
|
private readonly dynamic df;
|
|
|
|
public PandasTA() {
|
|
bars = new GBM_Feed(5000, 0.8, 0.0);
|
|
period = rnd.Next(28) + 3;
|
|
skip = period + 50;
|
|
digits = 8;
|
|
|
|
var pythonDLL = PythonLibrary.Locate();
|
|
Runtime.PythonDLL = pythonDLL;
|
|
PythonEngine.Initialize();
|
|
|
|
np = Py.Import("numpy");
|
|
pd = Py.Import("pandas");
|
|
ta = Py.Import("pandas_ta");
|
|
|
|
string[] cols = {"open", "high", "low", "close", "volume"};
|
|
var ary = new double[bars.Count, 5];
|
|
for (var i = 0; i < bars.Count; i++) {
|
|
ary[i, 0] = bars.Open[i].v;
|
|
ary[i, 1] = bars.High[i].v;
|
|
ary[i, 2] = bars.Low[i].v;
|
|
ary[i, 3] = bars.Close[i].v;
|
|
ary[i, 4] = bars.Volume[i].v;
|
|
}
|
|
|
|
df = ta.DataFrame(data: np.array(ary), index: np.array(bars.Close.t), columns: np.array(cols));
|
|
}
|
|
|
|
public void Dispose() {
|
|
Dispose(true);
|
|
PythonEngine.Shutdown();
|
|
GC.SuppressFinalize(this);
|
|
}
|
|
|
|
~PandasTA() {
|
|
Dispose(false);
|
|
}
|
|
|
|
protected virtual void Dispose(bool disposing) {
|
|
if (!disposed) {
|
|
disposed = true;
|
|
}
|
|
}
|
|
|
|
[Fact]
|
|
private void ADL() {
|
|
ADL_Series QL = new(bars);
|
|
var pta = df.ta.ad(high: df.high, low: df.low, close: df.close, volume: df.volume);
|
|
for (var i = QL.Length - 1; i > skip; i--) {
|
|
var QL_item = QL[i - 1].v;
|
|
var PanTA_item = (double) pta[i - 1];
|
|
Assert.InRange(PanTA_item! - QL_item, -Math.Exp(-digits), Math.Exp(-digits));
|
|
}
|
|
}
|
|
|
|
[Fact]
|
|
private void BBANDS() {
|
|
BBANDS_Series QL = new(bars.Close, period);
|
|
var pta = df.ta.bbands(close: df.close, length: period).to_numpy();
|
|
for (var i = QL.Length - 1; i > skip; i--) {
|
|
var QL_item = QL.Lower[i].v;
|
|
var PanTA_item = (double) pta[i][0]; //lower
|
|
Assert.InRange(PanTA_item! - QL_item, -Math.Exp(-digits), Math.Exp(-digits));
|
|
QL_item = QL.Mid[i].v;
|
|
PanTA_item = (double) pta[i][1]; //mid
|
|
Assert.InRange(PanTA_item! - QL_item, -Math.Exp(-digits), Math.Exp(-digits));
|
|
QL_item = QL.Upper[i].v;
|
|
PanTA_item = (double) pta[i][2]; //upper
|
|
Assert.InRange(PanTA_item! - QL_item, -Math.Exp(-digits), Math.Exp(-digits));
|
|
}
|
|
}
|
|
|
|
[Fact]
|
|
private void BIAS() {
|
|
BIAS_Series QL = new(bars.Close, period, false);
|
|
var pta = df.ta.bias(close: df.close, length: period);
|
|
for (var i = QL.Length - 1; i > skip; i--) {
|
|
var QL_item = QL[i - 1].v;
|
|
var PanTA_item = (double) pta[i - 1];
|
|
Assert.InRange(PanTA_item! - QL_item, -Math.Exp(-digits), Math.Exp(-digits));
|
|
}
|
|
}
|
|
|
|
[Fact]
|
|
private void CCI() {
|
|
CCI_Series QL = new(bars, period, false);
|
|
var pta = df.ta.cci(close: df.close, length: period);
|
|
for (var i = QL.Length - 1; i > skip; i--) {
|
|
var QL_item = QL[i - 1].v;
|
|
var PanTA_item = (double) pta[i - 1];
|
|
Assert.InRange(PanTA_item! - QL_item, -Math.Exp(-digits), Math.Exp(-digits));
|
|
}
|
|
}
|
|
|
|
[Fact]
|
|
private void DEMA() {
|
|
DEMA_Series QL = new(bars.Close, period, false);
|
|
var pta = df.ta.dema(close: df.close, length: period);
|
|
for (var i = QL.Length - 1; i > skip; i--) {
|
|
var QL_item = QL[i - 1].v;
|
|
var PanTA_item = (double) pta[i - 1];
|
|
Assert.InRange(PanTA_item! - QL_item, -Math.Exp(-digits), Math.Exp(-digits));
|
|
}
|
|
}
|
|
|
|
[Fact]
|
|
private void EMA() {
|
|
EMA_Series QL = new(bars.Close, period, false);
|
|
var pta = df.ta.ema(close: df.close, length: period);
|
|
for (var i = QL.Length - 1; i > skip; i--) {
|
|
var QL_item = QL[i - 1].v;
|
|
var PanTA_item = (double) pta[i - 1];
|
|
Assert.InRange(PanTA_item! - QL_item, -Math.Exp(-digits), Math.Exp(-digits));
|
|
}
|
|
}
|
|
|
|
[Fact]
|
|
private void ENTROPY() {
|
|
ENTROPY_Series QL = new(bars.Close, period, false);
|
|
var pta = df.ta.entropy(close: df.close, length: period);
|
|
for (var i = QL.Length - 1; i > skip; i--) {
|
|
var QL_item = QL[i - 1].v;
|
|
var PanTA_item = (double) pta[i - 1];
|
|
Assert.InRange(PanTA_item! - QL_item, -Math.Exp(-digits), Math.Exp(-digits));
|
|
}
|
|
}
|
|
|
|
[Fact]
|
|
private void HL2() {
|
|
var pta = df.ta.hl2(high: df.high, low: df.low);
|
|
for (var i = bars.HL2.Length - 1; i > skip; i--) {
|
|
var QL_item = bars.HL2[i - 1].v;
|
|
var PanTA_item = (double) pta[i - 1];
|
|
Assert.InRange(PanTA_item! - QL_item, -Math.Exp(-digits), Math.Exp(-digits));
|
|
}
|
|
}
|
|
|
|
[Fact]
|
|
private void HLC3() {
|
|
var pta = df.ta.hlc3(high: df.high, low: df.low, close: df.close);
|
|
for (var i = bars.HLC3.Length; i > skip; i--) {
|
|
var QL_item = bars.HLC3[i - 1].v;
|
|
var PanTA_item = (double) pta[i - 1];
|
|
Assert.InRange(PanTA_item! - QL_item, -Math.Exp(-digits), Math.Exp(-digits));
|
|
}
|
|
}
|
|
|
|
[Fact]
|
|
private void HMA() {
|
|
HMA_Series QL = new(bars.Close, period, false);
|
|
var pta = df.ta.hma(close: df.close, length: period);
|
|
for (var i = QL.Length - 1; i > skip; i--) {
|
|
var QL_item = QL[i - 1].v;
|
|
var PanTA_item = (double) pta[i - 1];
|
|
Assert.InRange(PanTA_item! - QL_item, -Math.Exp(-digits), Math.Exp(-digits));
|
|
}
|
|
}
|
|
|
|
[Fact]
|
|
private void KURTOSIS() {
|
|
KURTOSIS_Series QL = new(bars.Close, period, false);
|
|
var pta = df.ta.kurtosis(close: df.close, length: period);
|
|
for (var i = QL.Length - 1; i > skip; i--) {
|
|
var QL_item = QL[i - 1].v;
|
|
var PanTA_item = (double) pta[i - 1];
|
|
Assert.InRange(PanTA_item! - QL_item, -Math.Exp(-digits), Math.Exp(-digits));
|
|
}
|
|
}
|
|
|
|
[Fact]
|
|
private void MACD() {
|
|
MACD_Series QL = new(bars.Close, 26, 12, 9, false);
|
|
var pta = df.ta.macd(close: df.close).to_numpy();
|
|
for (var i = QL.Length - 1; i > skip; i--) {
|
|
var QL_item = QL[i - 1].v;
|
|
var PanTA_item = (double) pta[i - 1][0];
|
|
Assert.InRange(PanTA_item! - QL_item, -Math.Exp(-digits), Math.Exp(-digits));
|
|
QL_item = QL.Signal[i - 1].v;
|
|
PanTA_item = (double) pta[i - 1][2];
|
|
Assert.InRange(PanTA_item! - QL_item, -Math.Exp(-digits), Math.Exp(-digits));
|
|
}
|
|
}
|
|
|
|
[Fact]
|
|
private void MAD() {
|
|
MAD_Series QL = new(bars.Close, period, false);
|
|
var pta = df.ta.mad(close: df.close, length: period);
|
|
for (var i = QL.Length - 1; i > skip; i--) {
|
|
var QL_item = QL[i - 1].v;
|
|
var PanTA_item = (double) pta[i - 1];
|
|
Assert.InRange(PanTA_item! - QL_item, -Math.Exp(-digits), Math.Exp(-digits));
|
|
}
|
|
}
|
|
|
|
[Fact]
|
|
private void MEDIAN() {
|
|
MEDIAN_Series QL = new(bars.Close, period);
|
|
var pta = df.ta.median(close: df.close, length: period);
|
|
for (var i = QL.Length - 1; i > skip; i--) {
|
|
var QL_item = QL[i - 1].v;
|
|
var PanTA_item = (double) pta[i - 1];
|
|
Assert.InRange(PanTA_item! - QL_item, -Math.Exp(-digits), Math.Exp(-digits));
|
|
}
|
|
}
|
|
|
|
[Fact]
|
|
private void OBV() {
|
|
OBV_Series QL = new(bars);
|
|
var pta = df.ta.obv(close: df.close, volume: df.volume);
|
|
for (var i = QL.Length - 1; i > skip; i--) {
|
|
var QL_item = QL[i - 1].v;
|
|
var PanTA_item = (double) pta[i - 1];
|
|
Assert.InRange(PanTA_item! - QL_item, -Math.Exp(-digits), Math.Exp(-digits));
|
|
}
|
|
}
|
|
|
|
[Fact]
|
|
private void OHLC4() {
|
|
var pta = df.ta.ohlc4(open: df.open, high: df.high, low: df.low, close: df.close);
|
|
for (var i = bars.OHLC4.Length; i > skip; i--) {
|
|
var QL_item = bars.OHLC4[i - 1].v;
|
|
var PanTA_item = (double) pta[i - 1];
|
|
Assert.InRange(PanTA_item! - QL_item, -Math.Exp(-digits), Math.Exp(-digits));
|
|
}
|
|
}
|
|
|
|
[Fact]
|
|
private void SDEV() {
|
|
SDEV_Series QL = new(bars.Close, period, false);
|
|
var pta = df.ta.stdev(close: df.close, length: period, ddof: 0);
|
|
for (var i = QL.Length - 1; i > skip; i--) {
|
|
var QL_item = QL[i - 1].v;
|
|
var PanTA_item = (double) pta[i - 1];
|
|
Assert.InRange(PanTA_item! - QL_item, -Math.Exp(-digits), Math.Exp(-digits));
|
|
}
|
|
}
|
|
|
|
[Fact]
|
|
private void SMA() {
|
|
SMA_Series QL = new(bars.Close, period, false);
|
|
var pta = df.ta.sma(close: df.close, length: period);
|
|
for (var i = QL.Length - 1; i > skip; i--) {
|
|
var QL_item = QL[i - 1].v;
|
|
var PanTA_item = (double) pta[i - 1];
|
|
Assert.InRange(PanTA_item! - QL_item, -Math.Exp(-digits), Math.Exp(-digits));
|
|
}
|
|
}
|
|
|
|
[Fact]
|
|
private void SSDEV() {
|
|
SSDEV_Series QL = new(bars.Close, period, false);
|
|
var pta = df.ta.stdev(close: df.close, length: period, ddof: 1);
|
|
for (var i = QL.Length - 1; i > skip; i--) {
|
|
var QL_item = QL[i - 1].v;
|
|
var PanTA_item = (double) pta[i - 1];
|
|
Assert.InRange(PanTA_item! - QL_item, -Math.Exp(-digits), Math.Exp(-digits));
|
|
}
|
|
}
|
|
|
|
[Fact]
|
|
private void SVARIANCE() {
|
|
SVAR_Series QL = new(bars.Close, period);
|
|
var pta = df.ta.variance(close: df.close, length: period, ddof: 1);
|
|
for (var i = QL.Length - 1; i > skip; i--) {
|
|
var QL_item = QL[i - 1].v;
|
|
var PanTA_item = (double) pta[i - 1];
|
|
Assert.InRange(PanTA_item! - QL_item, -Math.Exp(-digits), Math.Exp(-digits));
|
|
}
|
|
}
|
|
|
|
[Fact]
|
|
private void TEMA() {
|
|
TEMA_Series QL = new(bars.Close, period, false);
|
|
var pta = df.ta.tema(close: df.close, length: period);
|
|
for (var i = QL.Length - 1; i > skip; i--) {
|
|
var QL_item = QL[i - 1].v;
|
|
var PanTA_item = (double) pta[i - 1];
|
|
Assert.InRange(PanTA_item! - QL_item, -Math.Exp(-digits), Math.Exp(-digits));
|
|
}
|
|
}
|
|
|
|
[Fact]
|
|
private void TR() {
|
|
TR_Series QL = new(bars);
|
|
var pta = df.ta.true_range(high: df.high, low: df.low, close: df.close);
|
|
for (var i = QL.Length - 1; i > skip; i--) {
|
|
var QL_item = QL[i - 1].v;
|
|
var PanTA_item = (double) pta[i - 1];
|
|
Assert.InRange(PanTA_item! - QL_item, -Math.Exp(-digits), Math.Exp(-digits));
|
|
}
|
|
}
|
|
|
|
[Fact]
|
|
private void TRIMA() {
|
|
// TODO: return length to variable length (period) when Pandas-TA fixes trima to calculate even periods right
|
|
TRIMA_Series QL = new(bars.Close, 11);
|
|
var pta = df.ta.trima(close: df.close, length: 11);
|
|
for (var i = QL.Length - 1; i > skip; i--) {
|
|
var QL_item = QL[i - 1].v;
|
|
var PanTA_item = (double) pta[i - 1];
|
|
Assert.InRange(PanTA_item! - QL_item, -Math.Exp(-digits), Math.Exp(-digits));
|
|
}
|
|
}
|
|
|
|
[Fact]
|
|
private void VARIANCE() {
|
|
VAR_Series QL = new(bars.Close, period);
|
|
var pta = df.ta.variance(close: df.close, length: period, ddof: 0);
|
|
for (var i = QL.Length - 1; i > skip; i--) {
|
|
var QL_item = QL[i - 1].v;
|
|
var PanTA_item = (double) pta[i - 1];
|
|
Assert.InRange(PanTA_item! - QL_item, -Math.Exp(-digits), Math.Exp(-digits));
|
|
}
|
|
}
|
|
|
|
[Fact]
|
|
private void WMA() {
|
|
WMA_Series QL = new(bars.Close, period, false);
|
|
var pta = df.ta.wma(close: df.close, length: period);
|
|
for (var i = QL.Length - 1; i > skip; i--) {
|
|
var QL_item = QL[i - 1].v;
|
|
var PanTA_item = (double) pta[i - 1];
|
|
Assert.InRange(PanTA_item! - QL_item, -Math.Exp(-digits), Math.Exp(-digits));
|
|
}
|
|
}
|
|
|
|
[Fact]
|
|
private void ZSCORE() {
|
|
ZSCORE_Series QL = new(bars.Close, period, false);
|
|
var pta = df.ta.zscore(close: df.close, length: period, ddof: 0);
|
|
for (var i = QL.Length - 1; i > skip; i--) {
|
|
var QL_item = QL[i - 1].v;
|
|
var PanTA_item = (double) pta[i - 1];
|
|
Assert.InRange(PanTA_item! - QL_item, -Math.Exp(-digits), Math.Exp(-digits));
|
|
}
|
|
}
|
|
}
|
|
|
|
public static class PythonLibrary {
|
|
public static string Locate() {
|
|
if (RuntimeInformation.IsOSPlatform(OSPlatform.Windows)) {
|
|
string[] paths = Environment.GetEnvironmentVariable("PATH")?.Split(';') ?? Array.Empty<string>();
|
|
foreach (string path in paths) {
|
|
string[] pythonDLLs = Directory.GetFiles(path, "python3*.dll");
|
|
if (pythonDLLs.Length > 0) {
|
|
foreach (string item in pythonDLLs) {
|
|
if (!item.EndsWith("python3.dll", StringComparison.OrdinalIgnoreCase)) {
|
|
return item;
|
|
}
|
|
}
|
|
|
|
}
|
|
}
|
|
throw new FileNotFoundException("Python library not found in PATH");
|
|
}
|
|
else if (RuntimeInformation.IsOSPlatform(OSPlatform.Linux)) {
|
|
return "/usr/lib/x86_64-linux-gnu/libpython3.10.so";
|
|
/*
|
|
List<string> pythonLibraries = new List<string>();
|
|
List<string> directoriesToSearch = new List<string> { "/home/runner/.local/lib" }; // Add more directories as needed
|
|
string filePattern = "libpython3.*.so";
|
|
SearchFiles(directoriesToSearch, filePattern, pythonLibraries);
|
|
|
|
if (pythonLibraries.Count > 0) {
|
|
return pythonLibraries[0];
|
|
}
|
|
else {
|
|
throw new FileNotFoundException("Python library not found");
|
|
}
|
|
*/
|
|
}
|
|
|
|
else if (RuntimeInformation.IsOSPlatform(OSPlatform.OSX)) {
|
|
throw new NotSupportedException("Not supported yet");
|
|
}
|
|
|
|
else { throw new NotSupportedException("Unsupported operating system"); }
|
|
}
|
|
static void SearchFiles(List<string> directoriesToSearch, string filePattern, List<string> foundFiles)
|
|
{
|
|
foreach (string directory in directoriesToSearch)
|
|
{
|
|
if (Directory.Exists(directory))
|
|
{
|
|
try
|
|
{
|
|
string[] files = Directory.GetFiles(directory, filePattern, SearchOption.AllDirectories);
|
|
foundFiles.AddRange(files);
|
|
}
|
|
catch (Exception e)
|
|
{
|
|
Console.WriteLine("Error searching in directory: " + directory + " - " + e.Message);
|
|
}
|
|
}
|
|
}
|
|
}
|
|
} |