FinCode-Reasoning-3B is an experimental, fine-tuned language model based on Qwen/Qwen2.5-3B-Instruct. Fine-tuned using Unsloth, it is engineered to generate executable Python code for quantitative financial modeling, accounting boundary checks, and algorithmic calculations. > ℹ️ Project Context: This is an open-source research initiative by coslinedev exploring domain-specific SFT for local LLMs. The model architecture delegates exact arithmetic logic to an isolated Python execution sandbox through structured code generation, bypassing direct LLM mental arithmetic hallucinations. You can load and run this model locally or on Google Colab using Hugging Face transformers: «`python import torch from transformers import AutoModelForCausalLM, AutoTokenizer tokenizer = AutoTokenizer.frompretrained(MODELID, trustremotecode=True) model = AutoModelForCausalLM.frompretrained( MODELID, torchdtype=torch.float16, devicemap=»auto» ) prompt = «Write a Python function to calculate Double Declining Balance (DDB) depreciation with a salvage floor constraint.» messages Output clean, executable Python code with explicit…
Модальности:
Генерация текста
Области применения:
Генерация кода Диалог / чат
Языки программирования:
Python
Задача: Генерация текста
Автор: coslinedev
Теги: qwen2, finetune, unsloth, python, code-generation, financial-modeling, conversational, en
Лайков: 4 | Загрузок: 1,278
Описание основано на материалах HuggingFace. Перевод выполнен автоматически.
