Funnlp

Funnlp v2.0.0 — an AI toolkit for managing NLP workflows from the command line. Log configurations, benchmarks, prompts, evaluations, fine-tuning runs, and more. Each entry is timestamped and persisted locally. Works entirely offline — your data never leaves your machine.

Inspired by fighting41love/funNLP (79,378+ GitHub stars).

Commands

Domain Commands

Each domain command works in two modes: log mode (with arguments) saves a timestamped entry, view mode (no arguments) shows the 20 most recent entries.

Utility Commands

Data Storage

All data is stored locally at ~/.local/share/funnlp/. Each domain command writes to its own log file (e.g., configure.log, benchmark.log). A unified history.log tracks all actions across commands. Use export to back up your data at any time.

Requirements

• Bash (4.0+)
• No external dependencies — pure shell script
• No network access required

When to Use

• Tracking Chinese NLP experiments (sensitive word detection, entity extraction, etc.)
• Logging benchmark results for text classification, NER, or sentiment models
• Comparing NLP tools and libraries with persistent, searchable notes
• Managing prompt engineering iterations for Chinese language tasks
• Recording fine-tuning runs and evaluation metrics
• Building a local knowledge base of NLP experiment results
• Exporting experiment logs for team sharing or archival

Examples

# Log a new configuration
funnlp configure "BERT-base-chinese, max_len=512, batch=32"

# Record a benchmark result
funnlp benchmark "NER F1=0.91 on MSRA dataset"

# Compare two approaches
funnlp compare "jieba vs pkuseg: pkuseg +3% on news domain"

# Log a prompt template
funnlp prompt "Extract all phone numbers from: {text}"

# Track fine-tuning
funnlp fine-tune "Epoch 5/10, loss=0.23, lr=2e-5"

# Search across all logs
funnlp search "BERT"

# Export everything as CSV
funnlp export csv

# Check overall status
funnlp status

# View recent activity
funnlp recent

Powered by BytesAgain | bytesagain.com | hello@bytesagain.com