Metadata-Version: 2.1
Name: iblm
Version: 0.3.66
Summary: Inductive-bias Learning
License: MIT
Author: toma tanaka
Author-email: ulti4929@gmail.com
Requires-Python: >=3.9,<3.12
Classifier: License :: OSI Approved :: MIT License
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.9
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Requires-Dist: anthropic (>=0.19.1,<0.20.0)
Requires-Dist: datasets (>=2.13.1,<3.0.0)
Requires-Dist: google-generativeai (>=0.3.2,<0.4.0)
Requires-Dist: jinja2 (>=3.1.0,<4.0.0)
Requires-Dist: numpy (>=1.21,<1.24)
Requires-Dist: openai (==1.3.7)
Requires-Dist: pandas (==1.4.0)
Requires-Dist: scikit-learn (>=1.2.2,<2.0.0)
Requires-Dist: tqdm (>=4.65.0,<5.0.0)
Description-Content-Type: text/markdown

# IBLM:Inductive-bias Learning Models
<div align="center">

[[ArXiv]](https://arxiv.org/abs/2308.09890)

</div>

- [What is IBL](#what-is-ibl)
- [How to Use](#how-to-use)
- [Examples](#examples)
- [Supported Models](#supported-models)
- [Contributor](#contributor)
- [Backstory](#backstory)



## What is IBL?
IBL (Inductive-bias Learning) is a new machine learning modeling method that uses LLM to infer the structure of the model itself from the data set and outputs it as Python code. The learned model (code model) can be used as a machine learning model to predict a new dataset.In this repository, you can try different learning methods with IBL.(Currently only binary classification with simple methods is available.)

![ibl](./images/ibl.png)


## How to Use

-  Installation and Import
```python
pip install iblm

import iblm
```

- Setting
  - OpenAI
```python
#
os.environ["OPENAI_API_KEY"] = "YOUR_API_KEY"

model = iblm.IBLModel(api_type="openai", model_name="gpt-4-0125-preview", objective="binary")
```

  - Azure OpenAI
```python
# AZURE_OPENAI_API
os.environ["AZURE_OPENAI_KEY"] = "xxx"
os.environ["AZURE_OPENAI_ENDPOINT"] = "xxx"
os.environ["OPENAI_API_VERSION"] = "xxx"

model = iblm.IBLModel(api_type="azure", model_name="gpt-4-0125-preview", objective="binary")
```

  - Google API
```python
os.environ["GOOGLE_API_KEY"] = "YOUR_API_KEY"
model = iblm.IBLModel(api_type="gemini", model_name="gemini-pro", objective="binary")
```

  - Anthropic API
```python
os.environ["ANTHROPIC_API_KEY"] = "YOUR_API_KEY"
model = iblm.IBLModel(api_type="", model_name="", objective="binary")
```

-  Model Learning
Currently, only small amounts of data can be executed.
```python
code_model = model.fit(x_train, y_train)

print(code_model)
```

-  Model Predictions
```python
y_proba = model.predict(x_test)
```

## Examples
Use the link below to try it out immediately on Google colab.
- Binary classification
    - Pseudo dataset:[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/fuyu-quant/IBLM/blob/main/examples/iblmodel/pseudodata.ipynb)
    - Moon dataset:[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/fuyu-quant/IBLM/blob/main/examples/iblmodel/iblmodel_moon.ipynb)


## Supported Models
- [OpenAI](https://platform.openai.com/api-keys)
  - gpt-4-0125-preview
  - gpt-3.5-turbo-0125
- [Azure OpenAI]()
  - gpt-4-0125-preview
  - gpt-3.5-turbo-0125
- [Google]()
  - gemini-pro
- [Anthropic]()
  - claude-3-opus-20240229
  - claude-3-sonnet-20240229


## Contributor
- [@t-ymbys](https://github.com/t-ymbys)
- [@cn47](https://github.com/cn47)


## Cite
If you find this repo helpful, please cite the following papers:
```
@article{tanaka2023inductive,
  title={Inductive-bias Learning: Generating Code Models with Large Language Model},
  author={Tanaka, Toma and Emoto, Naofumi and Yumibayashi, Tsukasa},
  journal={arXiv preprint arXiv:2308.09890},
  year={2023}
}
```

