Strateji türü

Makine öğrenmesi tabanlı işlem algoritmaları

Nicel finans için açık kaynak ML modelleri

Makine öğrenmesi tabanlı işlem algoritmaları, finansal zaman serisi verilerine istatistiksel öğrenme modellerini (gradient boosting, sinir ağları, Transformer'lar) uygulayarak fiyat hareketlerini tahmin eder veya işlem sinyalleri üretir. Bu açık kaynak uygulamalar hem regresyon (fiyat hedefi) hem de sınıflandırma (yön) formülasyonlarını kapsar.

13 algoritma2 kütüphane

Algoritma ağı

ML algoritmalarının kütüphaneler arasında nasıl bağlandığı

🧠ML Algorithms

🤖

Freqtrade9 algos

🔬

Microsoft Qlib6 algos

LightGBMRegressorintermediate

LightGBMClassifierintermediate

XGBoostRegressorintermediate

XGBoostClassifierintermediate

CatboostRegressorintermediate

PyTorchMLPRegressoradvanced

PyTorchTransformerRegressoradvanced

LGBModelintermediate

XGBModelintermediate

DNNModeladvanced

ALSTMadvanced

TFTModeladvanced

GATsadvanced

İşlem karar boru hattı

ML algoritmalarının bir işlem sisteminde nasıl birlikte çalıştığı

📊

Market Data Input

Price, volume, indicators

OHLCV price data

Technical indicators

Feature engineering

🧠

ML Signal Generation

Model predictions

LightGBM/XGBoost regression

Transformer sequence modeling

MLP classification

📈

Entry Decision

Signal threshold filtering

Confidence score > 0.6

Direction: Long/Short

📉

Exit Strategy

Profit taking & stop loss

Take profit (fixed % or trailing)

Stop loss (ATR-based)

🛡️

Risk Management

Position sizing & portfolio

Position sizing (Kelly criterion)

Portfolio rebalancing

Algoritma karşılaştırması

ML algoritmalarını temel boyutlarda karşılaştırın

Algoritma Karşılaştırma MatrisiDetayları genişletmek için bir sütuna tıklayın

Metrik	LightGBMRegressorFreqtrade	LightGBMClassifierFreqtrade	XGBoostRegressorFreqtrade	XGBoostClassifierFreqtrade	CatboostRegressorFreqtrade	PyTorchMLPRegressorFreqtrade	PyTorchTransformerRegressorFreqtrade	LGBModelQlib (Microsoft)	XGBModelQlib (Microsoft)	DNNModelQlib (Microsoft)	ALSTMQlib (Microsoft)	TFTModelQlib (Microsoft)	GATsQlib (Microsoft)
🎯Karmaşıklık	⭐⭐⭐intermediate	⭐⭐⭐intermediate	⭐⭐⭐intermediate	⭐⭐⭐intermediate	⭐⭐⭐intermediate	⭐⭐⭐⭐advanced	⭐⭐⭐⭐advanced	⭐⭐⭐intermediate	⭐⭐⭐intermediate	⭐⭐⭐⭐advanced	⭐⭐⭐⭐advanced	⭐⭐⭐⭐advanced	⭐⭐⭐⭐advanced
📈Tahmin Türü	Regresyon	Sınıflandırma	Regresyon	Sınıflandırma	Regresyon	Regresyon	Regresyon	Karışık	Karışık	Karışık	Dizi	Karışık	Karışık
⚡Eğitim Hızı	⚡⚡⚡	⚡⚡⚡	⚡⚡⚡	⚡⚡⚡	⚡⚡	⚡⚡	⚡	⚡⚡	⚡⚡	⚡⚡	⚡	⚡⚡	⚡⚡
🎯Doğruluk	📊📊📊	📊📊📊	📊📊📊	📊📊📊	📊📊📊	📊📊📊	📊📊📊📊	📊📊	📊📊	📊📊	📊📊📊	📊📊	📊📊📊📊
💡Şunun için en iyi	Tablo verisi	Tablo verisi	Tablo verisi	Tablo verisi	Genel amaçlı	Doğrusal olmayan desenler	Zaman serisi desenleri	Genel amaçlı	Genel amaçlı	Genel amaçlı	Sıralı veri	Genel amaçlı	Graf ilişkileri

Freqtrade

LightGBMRegressor

Freqtrade

Makine Öğrenmesiintermediate

Gradient boosting regression model for price movement prediction using LightGBM.

Hız⚡⚡⚡

Doğruluk📊📊📊

Temel Parametreler

n_estimators	1000	Number of boosting rounds
learning_rate	0.01	Step size shrinkage

Kaynak:freqai/prediction_models/LightGBMRegressor.py

LightGBMClassifier

Freqtrade

Makine Öğrenmesiintermediate

Gradient boosting classification model for directional prediction (up/down/neutral).

Hız⚡⚡⚡

Doğruluk📊📊📊

Temel Parametreler

n_estimators

1000

Number of boosting rounds

Kaynak:freqai/prediction_models/LightGBMClassifier.py

XGBoostRegressor

Freqtrade

Makine Öğrenmesiintermediate

XGBoost-based regression model for continuous value prediction.

Hız⚡⚡⚡

Doğruluk📊📊📊

Temel Parametreler

n_estimators	1000	Number of boosting rounds
max_depth	6	Maximum tree depth

Kaynak:freqai/prediction_models/XGBoostRegressor.py

XGBoostClassifier

Freqtrade

Makine Öğrenmesiintermediate

XGBoost-based classification model for directional prediction.

Hız⚡⚡⚡

Doğruluk📊📊📊

Temel Parametreler

n_estimators

1000

Number of boosting rounds

Kaynak:freqai/prediction_models/XGBoostClassifier.py

CatboostRegressor

Freqtrade

Makine Öğrenmesiintermediate

CatBoost gradient boosting model with native categorical feature support.

Hız⚡⚡

Doğruluk📊📊📊

Temel Parametreler

iterations

1000

Number of boosting iterations

Kaynak:freqai/prediction_models/CatboostRegressor.py

PyTorchMLPRegressor

Freqtrade

Makine Öğrenmesiadvanced

Multi-layer perceptron neural network for regression-based price prediction.

Hız⚡⚡

Doğruluk📊📊📊

Temel Parametreler

hidden_dim	128	Hidden layer dimension
dropout_percent	0.2	Dropout rate

Kaynak:freqai/prediction_models/PyTorchMLPRegressor.py

PyTorchTransformerRegressor

Freqtrade

Makine Öğrenmesiadvanced

Transformer architecture for time-series regression using self-attention mechanism.

Hız⚡

Doğruluk📊📊📊📊

Temel Parametreler

num_heads	8	Number of attention heads
num_layers	2	Number of transformer layers

Kaynak:freqai/prediction_models/PyTorchTransformerRegressor.py

Qlib (Microsoft)

LGBModel

Qlib (Microsoft)

Makine Öğrenmesiintermediate

LightGBM model for stock return prediction using technical and fundamental features.

Hız⚡⚡

Doğruluk📊📊📊

Temel Parametreler

num_leaves	31	Maximum number of leaves
learning_rate	0.1	Boosting learning rate

Kaynak:qlib/contrib/model/gbdt.py

XGBModel

Qlib (Microsoft)

Makine Öğrenmesiintermediate

XGBoost model for stock return prediction.

Hız⚡⚡

Doğruluk📊📊📊

Kaynak:qlib/contrib/model/xgboost.py

DNNModel

Qlib (Microsoft)

Makine Öğrenmesiadvanced

Deep neural network for nonlinear feature extraction and return prediction.

Hız⚡⚡

Doğruluk📊📊📊

Temel Parametreler

hidden_size

256

Hidden layer size

Kaynak:qlib/contrib/model/pytorch_nn.py

ALSTM

Qlib (Microsoft)

Makine Öğrenmesiadvanced

Attention-based LSTM for sequential stock data modeling with attention mechanism.

Hız⚡

Doğruluk📊📊📊

Temel Parametreler

hidden_size	64	LSTM hidden size
num_layers	2	Number of LSTM layers

Kaynak:qlib/contrib/model/pytorch_alstm.py

TFTModel

Qlib (Microsoft)

Makine Öğrenmesiadvanced

Temporal Fusion Transformer combining static and temporal features for multi-horizon prediction.

Hız⚡⚡

Doğruluk📊📊📊

Kaynak:qlib/contrib/model/pytorch_tft.py

GATs

Qlib (Microsoft)

Makine Öğrenmesiadvanced

Graph Attention Networks modeling stock relationships for cross-stock prediction.

Hız⚡⚡

Doğruluk📊📊📊

Kaynak:qlib/contrib/model/pytorch_gats.py

Makine öğrenmesi tabanlı işlem algoritmaları, algoritma referansı

LightGBMRegressor (Freqtrade): Gradient boosting regression model for price movement prediction using LightGBM. Temel parametreler: n_estimators (Number of boosting rounds), learning_rate (Step size shrinkage).Kaynak: https://github.com/freqtrade/freqtrade/blob/develop/freqai/prediction_models/LightGBMRegressor.py.
LightGBMClassifier (Freqtrade): Gradient boosting classification model for directional prediction (up/down/neutral). Temel parametreler: n_estimators (Number of boosting rounds).Kaynak: https://github.com/freqtrade/freqtrade/blob/develop/freqai/prediction_models/LightGBMClassifier.py.
XGBoostRegressor (Freqtrade): XGBoost-based regression model for continuous value prediction. Temel parametreler: n_estimators (Number of boosting rounds), max_depth (Maximum tree depth).Kaynak: https://github.com/freqtrade/freqtrade/blob/develop/freqai/prediction_models/XGBoostRegressor.py.
XGBoostClassifier (Freqtrade): XGBoost-based classification model for directional prediction. Temel parametreler: n_estimators (Number of boosting rounds).Kaynak: https://github.com/freqtrade/freqtrade/blob/develop/freqai/prediction_models/XGBoostClassifier.py.
CatboostRegressor (Freqtrade): CatBoost gradient boosting model with native categorical feature support. Temel parametreler: iterations (Number of boosting iterations).Kaynak: https://github.com/freqtrade/freqtrade/blob/develop/freqai/prediction_models/CatboostRegressor.py.
PyTorchMLPRegressor (Freqtrade): Multi-layer perceptron neural network for regression-based price prediction. Temel parametreler: hidden_dim (Hidden layer dimension), dropout_percent (Dropout rate).Kaynak: https://github.com/freqtrade/freqtrade/blob/develop/freqai/prediction_models/PyTorchMLPRegressor.py.
PyTorchTransformerRegressor (Freqtrade): Transformer architecture for time-series regression using self-attention mechanism. Temel parametreler: num_heads (Number of attention heads), num_layers (Number of transformer layers).Kaynak: https://github.com/freqtrade/freqtrade/blob/develop/freqai/prediction_models/PyTorchTransformerRegressor.py.
LGBModel (Qlib (Microsoft)): LightGBM model for stock return prediction using technical and fundamental features. Temel parametreler: num_leaves (Maximum number of leaves), learning_rate (Boosting learning rate).Kaynak: https://github.com/microsoft/qlib/blob/main/qlib/contrib/model/gbdt.py.
XGBModel (Qlib (Microsoft)): XGBoost model for stock return prediction. Kaynak: https://github.com/microsoft/qlib/blob/main/qlib/contrib/model/xgboost.py.
DNNModel (Qlib (Microsoft)): Deep neural network for nonlinear feature extraction and return prediction. Temel parametreler: hidden_size (Hidden layer size).Kaynak: https://github.com/microsoft/qlib/blob/main/qlib/contrib/model/pytorch_nn.py.
ALSTM (Qlib (Microsoft)): Attention-based LSTM for sequential stock data modeling with attention mechanism. Temel parametreler: hidden_size (LSTM hidden size), num_layers (Number of LSTM layers).Kaynak: https://github.com/microsoft/qlib/blob/main/qlib/contrib/model/pytorch_alstm.py.
TFTModel (Qlib (Microsoft)): Temporal Fusion Transformer combining static and temporal features for multi-horizon prediction. Kaynak: https://github.com/microsoft/qlib/blob/main/qlib/contrib/model/pytorch_tft.py.
GATs (Qlib (Microsoft)): Graph Attention Networks modeling stock relationships for cross-stock prediction. Kaynak: https://github.com/microsoft/qlib/blob/main/qlib/contrib/model/pytorch_gats.py.