skills.rest
bianhaifeng789-hue
bianhaifeng789-hueCOMMUNITY
Explore all skills by this author →
GitHub

train-fasttext

Train FastText supervised text classification models within accuracy and size constraints.

Updated Apr 14, 2026
Search Tags:#quantization#machine-learning#hyperparameter-tuning#fasttext#text-classification#model-size
One-click install
npx skills add https://github.com/bianhaifeng789-hue/openclaw-config --skill train-fasttext-bianhaifeng789-hue
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: train-fasttext
Source: https://github.com/bianhaifeng789-hue/openclaw-config/tree/main/skills/tb2/train-fasttext
Command: npx skills add https://github.com/bianhaifeng789-hue/openclaw-config --skill train-fasttext-bianhaifeng789-hue

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Training FastText supervised text classification models while balancing accuracy and model size, and providing hyperparameter tuning guidance for compact deployments.

Core Features & Use Cases

  • Phase 1: Quick Exploration using a small data subset to identify promising parameter ranges (dim, epoch, lr, wordNgrams, minCount, bucket) with a focus on speed and iterative feedback.
  • Phase 2: Systematic Parameter Search to refine configurations for accuracy vs. size, including adjustments to dim, epoch, lr, minCount, bucket, and n-grams.
  • Phase 3: Full Training and Evaluation, including comparing non-quantized and quantized models and applying quantization only when needed to meet size constraints.
  • Quantization Considerations: Use quantization as a last resort if size targets are not met by parameter tuning, with awareness of estimated accuracy loss.

Quick Start

Train a FastText supervised model on a small subset of your data to quickly identify parameter ranges that meet your accuracy and size goals.

Frequently Asked Questions about train-fasttext

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

FAQPage Schema
How do I train a FastText model with compact accuracy for limited computation budgets?▼

To train FastText models with compact accuracy, you iteratively tune hyperparameters like dim, epoch, lr, wordNgrams, minCount, and bucket to balance accuracy and model size for constrained environments.

What is the best way to reduce FastText model size without losing too much accuracy?▼

Reducing FastText model size involves systematic parameter search across dim, bucket, and n-grams. If size targets remain unmet, apply quantization as a last resort while accounting for estimated accuracy loss.

How do I tune FastText hyperparameters for text classification efficiently?▼

Efficient FastText hyperparameter tuning uses a small data subset for quick exploration of promising ranges, followed by systematic parameter search to refine configurations for accuracy versus size.

When should I use quantization in FastText supervised text classification?▼

Use quantization in FastText supervised text classification only when parameter tuning fails to meet predefined size constraints, applying it as a final step with awareness of the accuracy trade-off.

Can I evaluate FastText quantized models against non-quantized models during training?▼

Yes, you can evaluate FastText quantized models against non-quantized models during the full training and evaluation phase to compare accuracy and determine if size constraints are met.

Related Skills

Explore More Data & Analytics Skills→
lazyFrogLOLlazyFrogLOL
127

train-fasttext

Guide FastText supervised text classification training under accuracy and size constraints.

Community
Intermediate
samuelmukotisamuelmukoti

serving-llms-vllm

Deploy LLMs with vLLM using PagedAttention and continuous batching.

Community
Advanced
eyadsibaieyadsibai
7

huggingface-tokenizers

Trains custom BPE and WordPiece tokenizers using HuggingFace tokenizers library.

Community
Intermediate
GuitarAlchemistGuitarAlchemist

ix-supervised

Train and evaluate supervised machine learning models in Rust.

Official
Advanced