M

machine learning

ModalLens is a research prototype for exploring and refining modal theories. The upcoming release will combine three layers: formal evidence, structural and learned analysis, and natural-language explanation.

Using Isabelle/HOL and Nitpick, it enumerates finite models and countermodels through blocking axioms, while Leo-III contributes proof evidence via TPTP/THF. The models are visualized as graphs and analysed using structural features, graphlets, clustering, and learned representations of heterogeneous semantic graphs to identify recurring patterns and differences between model families.

Selected structural patterns can then be translated into candidate refinement axioms that exclude the corresponding configurations. These axioms can be adopted automatically or with user input, followed by renewed model enumeration and analysis to examine the effects of each refinement.

LLM-generated explanations draw on the formal and structural evidence to explain findings and proposed refinements. Recorded provenance, reproducible runs, and visual exports make the process inspectable and suitable for experimental evaluation.

Hybrid cloud-edge ML system for predictive rain control with automated retraining, monitoring, and Raspberry Pi hardware actuation.

A landcover classification tool based for humans. Classifier does "traditional" supervised and unsupervised learning. Image segmentation and soon also object detection

$ mldev | is a data science experiment automation and reproducibility toolkit.

check our experiment templates: https://gitlab.com/mlrep

Trading Bot – Algorithmic Crypto Trading with AI Integration

This project is a powerful algorithmic trading bot for cryptocurrency markets. It combines traditional technical analysis with modern machine learning to generate accurate and intelligent trading decisions.

Key Features:

Candlestick Pattern Detection: Identifies classic reversal patterns such as Hammer, Doji, Engulfing, Shooting Star, and complex formations like triangle patterns.

Technical Indicators: Includes standard indicators (RSI, MACD, Moving Averages, Bollinger Bands) and advanced tools like Ichimoku Clouds, SuperTrend, Fibonacci Retracements, and more.

Machine Learning Integration: Uses LSTM-based models for time-series forecasting and momentum strategies, combined with indicator signals through weighted evaluation.

Dynamic Signal Weighting: Customizable signal weighting for patterns, indicators, and ML predictions with automatic adjustments to market volatility.

Trade Execution Engine: Supports long/short positions with stop-loss, take-profit, and trailing stop features. Automatically includes fees and tax deductions in profit calculations.

Backtesting & Debugging: Simulates strategies on historical data with detailed equity/value curve visualization and comprehensive debug logs.

Robust Error Handling: Detects and logs data inconsistencies, index errors, and processing issues to ensure stability.

Modular architecture with key components such as TraderBot, SignalHandler, PatternManager, IndicatorManager, MLModelHandler, SequenceManager, DataAPI, and CryptoCurrency. Additional support provided by PatternCalculator, IndicatorCalculator, and DataProcessing.

Version: V1.3.0.0 | GUI: V1.0.0 Author: Marian Seeger – info@seegersoftwaredevelopment.de

Hybrid Filtering with Spark MLlib

💬 Epic prompts to turbo-charge your LLM chatbots.

Bahn-Vorhersage - The best Train Delay Prediction System.

Your very own Assistant. Because you deserve it.

Toolchain to translate the Navajo language into American English

Provides biomedical plotting archetypes fully interoperable with the matplotlib API.

Simple neural network to recognise handwritten digits. Built from scratch with NumPy (and without PyTorch or other ML libraries). Trained on MNIST.

This is an advance IA trading platform that will focus on forex trading

Convenient training of linear models

Extract video representations (semantic, geometric, deep features) for the frames of any video.

Analysis of Kilter Board data, along with predictive models for V-grades based on holds and angle.

Analysis of Tension Board 2 data, along with predictive models for V-grades based on holds and angle.

RECOSIM is a program for recombination detection based on shared derived states of clustered synapomorphic transformations extracted from a phylogenomic analysis.

Découverte de ce monde (Data / ML) via un projet perso. Basés sur des relevés météo de différentes sources, avec des outils comme Pandas, Dask, Spark, Polars, ..., du ML et du DL. Une couche de visualisation via PowerBI.