skills.rest
Donaldshen27
Donaldshen27COMMUNITY
Explore all skills by this author →
GitHub

x-trend-architecture

Implement attention-based time-series models with LSTM encoders and cross-attention.

Updated Nov 14, 2025
Search Tags:#time-series#few-shot#attention#interpretability#x-trend#encoder-decoder#lstm
One-click install
npx skills add https://github.com/Donaldshen27/xtrend-vanilla --skill x-trend-architecture
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: x-trend-architecture
Source: https://github.com/Donaldshen27/xtrend-vanilla/tree/main/.claude/skills/x-trend-architecture
Command: npx skills add https://github.com/Donaldshen27/xtrend-vanilla --skill x-trend-architecture

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Provides a complete blueprint for implementing the X-Trend architecture, enabling practitioners to build cross-attentive time-series models for trend-following, forecasting, and interpretable trading signals.

Core Features & Use Cases

  • Input features: 8-dimensional vectors combining normalized returns and MACD indicators for robust temporal modeling.
  • VSN & embeddings: Feature selection networks and asset embeddings to handle multi-asset regimes and few-shot learning.
  • Attention & encoder-decoder: Self-attention over context, cross-attention between target and context, plus an encoder-decoder structure for prediction and position forecasting.
  • Use Case: Build an end-to-end trading model that predicts positions and forecasts distributions with interpretable attention patterns.

Quick Start

Initialize the X-Trend model and begin training using episodic learning as described in TRAINING.md.

Frequently Asked Questions about x-trend-architecture

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I build an attention-based time-series model for multi-asset trading?▼

Build attention-based time-series models using an encoder-decoder structure with LSTM encoders and multi-head cross-attention to generate interpretable predictions and position forecasts across multiple financial assets.

What is cross-attention used for in financial time-series forecasting?▼

Cross-attention in financial time-series forecasting aligns target asset queries with context sequences, enabling the model to capture multi-asset regime shifts and generate interpretable trend-following trading signals.

How do I handle few-shot learning for new assets in time-series forecasting?▼

Handle few-shot learning for new assets by applying Variable Selection Networks and entity embeddings, which adapt the model to diverse multi-asset regimes with limited historical data during episodic training.

Can I use MACD indicators as input features for an LSTM encoder-decoder model?▼

Use MACD indicators combined with normalized returns to form 8-dimensional input vectors, providing robust temporal modeling for the LSTM encoder-decoder architecture to predict asset positions.

How do I train a cross-attentive time-series model for trend-following strategies?▼

Train cross-attentive time-series models using episodic learning guidelines, iteratively updating multi-head attention weights and variable selection networks to optimize trend-following signal generation.

Does this architecture support interpretable predictions across multiple assets?▼

This architecture supports interpretable predictions across multiple assets by leveraging self-attention over context and cross-attention between target and context, producing transparent attention patterns for trading signals.

Related Skills

Explore More Data & Analytics Skills→
mahmoud20138mahmoud20138
10

ml-trading

Identify and optimize machine learning models for forecasting returns, regimes, and trading signals.

Community
Advanced
mrzhao1277891mrzhao1277891

Trading Signal Expert

Generate executable cryptocurrency trading signals from market analysis.

Community
Advanced
xchang1121xchang1121
6

triton-ascend-attention

Optimize attention operators on Ascend using Triton with QKV tiling and online softmax.

Community
Advanced
brainbytes-devbrainbytes-dev
10

deep-learning-trading

Forecast financial time series price movements using LSTM, Transformer, or CNN-1D deep learning models.

Community
Advanced