The open-source model tinymistral-v2-pycoder-instruct-248m - Free support for instruction-based Python code generation

Tinymistral V2 Pycoder Instruct 248m

Developed by jtatman

A Python-specific code generation model developed based on TinyMistral-248M-v2-Instruct, supporting instruction-formatted Python code generation

Large Language Model

Transformers

OtherOpen Source License:Apache-2.0 #Python code generation #Instruction fine-tuning #Lightweight model

Downloads 16

Release Time : 1/21/2024

Model Overview

This model is an instruction-tuned model optimized for Python code generation, using the ChatML format for interaction. It can generate Python code of varying quality. The model is in active development.

Model Features

Python-specific optimization

Specially fine-tuned for Python code generation tasks

Instruction interaction format

Supports ChatML-formatted instruction interaction

Lightweight model

A lightweight model with 248M parameters, suitable for resource-constrained environments

Model Capabilities

Python code generation

Instruction response

Function implementation

Use Cases

Code development

Function implementation

Generate Python function implementations based on instructions

Produces executable Python code

Code completion

Complete Python code snippets based on context

Generates context-appropriate code completions

🚀 Model Card for tinymistral-v2-pycoder-instruct-248m

This model card is for tinymistral-v2-pycoder-instruct, a Python-specific code generation model built on top of Locutusque/TinyMistral-248M-v2-Instruct. It aims to generate Python code effectively.

✨ Features

Generate Python code with varying quality using the instruct format.
Follow the ChatML format for interaction.

📦 Installation

No installation steps are provided in the original document, so this section is skipped.

💻 Usage Examples

Basic Usage

Use the following prompt format to start using the model:

<|im_start|>user 
Write a function for multiplying two numbers, from variables 'a' and 'b'.<|im_end|>
<|im_start|>assistant

Advanced Usage

The model can potentially be fine - tuned with a more comprehensive dataset. Experiments are currently in progress.

📚 Documentation

Model Details

This instruct model adheres to the original in using the ChatML format. An empty prompt will return various information from the base model, while using the instruct format will yield Python code of different quality levels.

Developed by: [Locutusque and M4ai]
Funded by: [Lint from a corner pocket]
Shared by: jtatman
Model type: MistralForCausalLM
License: [MIT]
Finetuned from model: Locutusque/TinyMistral-248M-v2

Uses

The main use of this model is to generate Python code. It may also be fine - tuned with a more comprehensive dataset in future experiments.

Training Details

Training Data

The model is trained on custom - formatted existing Python data from the following sources:

Training Procedure

The training is repeated according to the compute budget.

Preprocessing

The data is converted to the alpaca/instruct format.

Training Hyperparameters

Training regime: fp16, merge of parameter fine - tune adapters when necessary and helpful.

Evaluation

Metrics

The latest metrics are as follows:

Property	Details
epoch	4.87
global_step	220
learning_rate	0.00006713780918727916
loss	2.3736

📄 License

The model is licensed under the [MIT] license.

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご