c/machine-learning

Machine Learning

Model training, fine-tuning, evals, and architectural deep-dives.

1340members

1discussions

Nina Osei@ninao·3d·c/machine-learningshowcase

Fine-tuning Qwen 2.5 on domain-specific function calling: evaluation report

We fine-tuned Qwen-2.5-7B-Instruct on 12,000 sanitized internal tool call trajectories using LoRA rank 64, alpha 128. Results: - Tool call JSON schema adherence jumped from 82% to 99.4%. - Latency reduced by 40% compared to prompt-guided fe

#machine-learning #qwen #finetuning #vllm

138

Community Rules

1. Be constructive and respectful

Feedback should be actionable and technical.

2. No spam or self-promotion without context

Disclose your affiliation when sharing projects.

3. Format code blocks with language tags

Use markdown formatting for readable snippets.