Skip to main content

CLI tool for interacting with the Felafax platform; fine-tuning and inference on non-NVIDIA chipsets.

Project description

Felafax CLI

A command-line interface for fine-tuning and managing large language models.

Installation

pip install felafax-cli

Quick Start

  1. Login with your authentication token:
felafax-cli auth login --token <your-token>
  1. Initialize a fine-tuning configuration:
felafax-cli tune init-config
  1. Upload your training data:
felafax-cli files upload path/to/data.jsonl
  1. Start fine-tuning:
felafax-cli tune start --model <model_name> --config config.yml --dataset <dataset_id>

Available Commands

Authentication

# Login with your token
felafax-cli auth login --token <your-token> [--force]

Model Management

# List available fine-tuned models
felafax-cli model list

# Start an interactive chat session with a model
felafax-cli model chat <model_id> [--system-prompt "Your custom prompt"]

# Get detailed information about a model
felafax-cli model info <model_id>

# Delete a fine-tuned model
felafax-cli model delete <model_id>

Training Management

# Initialize a new config file
felafax-cli tune init-config

# Start a new fine-tuning job
felafax-cli tune start \
    --model <model_name> \
    --config path/to/config.yml \
    --dataset <dataset_id>

# List all training jobs
felafax-cli tune list

# Check job status
felafax-cli tune status --job-id <id>

# Stop a running job
felafax-cli tune stop --job-id <id>

File Management

# List files in storage
felafax-cli files list [--prefix <prefix>] [--limit <number>]

# Upload training data
felafax-cli files upload <file_path>

# Delete a file
felafax-cli files delete <file_path>

Configuration Example

When you run felafax-cli tune init-config, it creates a YAML file with the following structure:

hyperparameters:
  learning_rate: 1.0e-05
  batch_size: 32
  n_epochs: 4
  warmup_ratio: 0.0
lora:
  enabled: false
  r: 8
  alpha: 8
  dropout: 0.0

Support

For more information, visit our documentation at https://docs.felafax.ai

License

MIT License

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

felafax_cli-0.1.8.tar.gz (58.8 kB view details)

Uploaded Source

Built Distribution

felafax_cli-0.1.8-py3-none-any.whl (45.9 kB view details)

Uploaded Python 3

File details

Details for the file felafax_cli-0.1.8.tar.gz.

File metadata

  • Download URL: felafax_cli-0.1.8.tar.gz
  • Upload date:
  • Size: 58.8 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/5.1.1 CPython/3.13.0

File hashes

Hashes for felafax_cli-0.1.8.tar.gz
Algorithm Hash digest
SHA256 97ac98bf607e621998b7a8fa4b40faa7a6aedab1e9ba774cc4951beff10de82c
MD5 2a9684ae03dfecb8c3088f7e2dddd07f
BLAKE2b-256 c28328366e2540b4564256cfc8ff174209972408b0b97b117eb775519c271463

See more details on using hashes here.

File details

Details for the file felafax_cli-0.1.8-py3-none-any.whl.

File metadata

  • Download URL: felafax_cli-0.1.8-py3-none-any.whl
  • Upload date:
  • Size: 45.9 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/5.1.1 CPython/3.13.0

File hashes

Hashes for felafax_cli-0.1.8-py3-none-any.whl
Algorithm Hash digest
SHA256 14be794a21258f265dca15cb039c32bce4ed0ab7ac0045bee38d796d2faf5c83
MD5 67d967c79a531f4c73b79c29ae7794d0
BLAKE2b-256 4a66a21f96083c1ae40a2207d34d7a1d7c9c409edc9971203ebdbe67abbf0c54

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page