# 💠Future Frame
* This Python package allows you to interact with pre-trained foundation models for tabular data.
* Easily fine-tune them on your classification and regression use cases in a single line of code.
* Interested in what we're building? Join our [waitlist](https://futureframe.ai/).
## Installation
1. Install Future Frame with `pip` – more details on our [PyPI page](https://pypi.org/project/futureframe/).
```bash
pip install futureframe
```
2. Download model weights [here](https://drive.google.com/drive/folders/1-SVab4cv3nLaUJjyOlscKP_OxOBbs_4e?usp=sharing) and store the `weights` folder in your working directory.
## Quick Start
Use Future Frame to fine-tune a pre-trained foundation model on a classification task.
```python
# Import standard libraries
import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn.metrics import roc_auc_score
# Import Future Frame
import futureframe as ff
# Import data
dataset_name = "tests/data/churn.csv"
target_variable = "Churn"
df = pd.read_csv(dataset_name)
# Split data
X, y = df.drop(columns=[target_variable]), df[target_variable]
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=42)
# Fine-tune a pre-trained classifier with Future Frame
model = ff.models.CM2Classifier()
model.finetune(X_train, y_train)
# Make predictions with Future Frame
y_pred = model.predict(X_test)
# Evaluate your model
auc = roc_auc_score(y_test, y_pred)
print(f"AUC: {auc:0.2f}")
```
## Models
| Model Name | Paper Title | Paper | GitHub |
| ---------- | ---------------------------------------------------------- | --------------------------------------------------- | -------------------------------------- |
| CM2 | Towards Cross-Table Masked Pretraining for Web Data Mining | [Ye et al., 2024](https://arxiv.org/abs/2307.04308) | [Link](https://github.com/Chao-Ye/CM2) |
More foundation models will be integrated into the library soon. Stay stuned by joining our [waitlist](https://futureframe.ai/)!
## Links
* [Future Frame Official Website](https://futureframe.ai/)
* [`futureframe` PyPI Page](https://pypi.python.org/pypi/futureframe)
* [`futureframe` GitHub Repository](https://github.com/futureframeai/futureframe)
* Documentation: coming soon!
## Contributing
* We are currently under heavy development.
* If you'd like to contribute, please send us an email at <i>eduardo(at)futureframe.ai</i>.
* To report a bug, please write an [issue](https://github.com/futureframeai/futureframe/issues/new).
Raw data
{
"_id": null,
"home_page": "https://futureframe.ai",
"name": "futureframe",
"maintainer": "Eduardo Dadalto",
"docs_url": null,
"requires_python": "<4.0,>=3.10",
"maintainer_email": "eduardo@futureframe.ai",
"keywords": "data science, foundation models, tabular data",
"author": "Future Frame",
"author_email": null,
"download_url": "https://files.pythonhosted.org/packages/bf/1d/55b85573644a5dcdc7ae7c94d25374f1c0b3b440fafa5c103b3ff663e002/futureframe-0.1.8.tar.gz",
"platform": null,
"description": "# \ud83d\udca0 Future Frame\n\n* This Python package allows you to interact with pre-trained foundation models for tabular data.\n* Easily fine-tune them on your classification and regression use cases in a single line of code.\n* Interested in what we're building? Join our [waitlist](https://futureframe.ai/).\n\n## Installation\n\n1. Install Future Frame with `pip` \u2013 more details on our [PyPI page](https://pypi.org/project/futureframe/).\n\n```bash\npip install futureframe\n```\n\n2. Download model weights [here](https://drive.google.com/drive/folders/1-SVab4cv3nLaUJjyOlscKP_OxOBbs_4e?usp=sharing) and store the `weights` folder in your working directory.\n\n## Quick Start\n\nUse Future Frame to fine-tune a pre-trained foundation model on a classification task.\n\n```python\n# Import standard libraries\nimport pandas as pd\nfrom sklearn.model_selection import train_test_split\nfrom sklearn.metrics import roc_auc_score\n\n# Import Future Frame\nimport futureframe as ff\n\n# Import data\ndataset_name = \"tests/data/churn.csv\"\ntarget_variable = \"Churn\"\ndf = pd.read_csv(dataset_name)\n\n# Split data\nX, y = df.drop(columns=[target_variable]), df[target_variable]\nX_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=42)\n\n# Fine-tune a pre-trained classifier with Future Frame\nmodel = ff.models.CM2Classifier()\nmodel.finetune(X_train, y_train)\n\n# Make predictions with Future Frame\ny_pred = model.predict(X_test)\n\n# Evaluate your model\nauc = roc_auc_score(y_test, y_pred)\nprint(f\"AUC: {auc:0.2f}\")\n```\n\n## Models\n\n| Model Name | Paper Title | Paper | GitHub |\n| ---------- | ---------------------------------------------------------- | --------------------------------------------------- | -------------------------------------- |\n| CM2 | Towards Cross-Table Masked Pretraining for Web Data Mining | [Ye et al., 2024](https://arxiv.org/abs/2307.04308) | [Link](https://github.com/Chao-Ye/CM2) |\n\nMore foundation models will be integrated into the library soon. Stay stuned by joining our [waitlist](https://futureframe.ai/)!\n\n## Links\n\n* [Future Frame Official Website](https://futureframe.ai/)\n* [`futureframe` PyPI Page](https://pypi.python.org/pypi/futureframe)\n* [`futureframe` GitHub Repository](https://github.com/futureframeai/futureframe)\n* Documentation: coming soon!\n\n## Contributing\n\n* We are currently under heavy development.\n* If you'd like to contribute, please send us an email at <i>eduardo(at)futureframe.ai</i>.\n* To report a bug, please write an [issue](https://github.com/futureframeai/futureframe/issues/new).\n",
"bugtrack_url": null,
"license": null,
"summary": "Accelerate your data science workflow from months to days with foundation models for tabular data.",
"version": "0.1.8",
"project_urls": {
"Documentation": "https://futureframe.ai/docs",
"Homepage": "https://futureframe.ai",
"Repository": "https://github.com/futureframeai/futureframe"
},
"split_keywords": [
"data science",
" foundation models",
" tabular data"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "da11b62cd8efe857e9342689dec3d77c5b88ec22955e9438bd9ff43c2123a65a",
"md5": "c9a2f04a43e4c87d3da970a464899482",
"sha256": "1c9dde33bed343e89c8de6a537be8e985b9e9e7938eec7ebb76e3a3b3add46b2"
},
"downloads": -1,
"filename": "futureframe-0.1.8-py3-none-any.whl",
"has_sig": false,
"md5_digest": "c9a2f04a43e4c87d3da970a464899482",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": "<4.0,>=3.10",
"size": 32719,
"upload_time": "2024-06-20T09:40:02",
"upload_time_iso_8601": "2024-06-20T09:40:02.391339Z",
"url": "https://files.pythonhosted.org/packages/da/11/b62cd8efe857e9342689dec3d77c5b88ec22955e9438bd9ff43c2123a65a/futureframe-0.1.8-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "bf1d55b85573644a5dcdc7ae7c94d25374f1c0b3b440fafa5c103b3ff663e002",
"md5": "e1c4680c88b3d2f12cbe3514c4ee83ca",
"sha256": "207e0fd12fbfebf4fa58dc7c7861b1f112658bf1decda799a7cbcf5bc35012b2"
},
"downloads": -1,
"filename": "futureframe-0.1.8.tar.gz",
"has_sig": false,
"md5_digest": "e1c4680c88b3d2f12cbe3514c4ee83ca",
"packagetype": "sdist",
"python_version": "source",
"requires_python": "<4.0,>=3.10",
"size": 30076,
"upload_time": "2024-06-20T09:40:11",
"upload_time_iso_8601": "2024-06-20T09:40:11.600450Z",
"url": "https://files.pythonhosted.org/packages/bf/1d/55b85573644a5dcdc7ae7c94d25374f1c0b3b440fafa5c103b3ff663e002/futureframe-0.1.8.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-06-20 09:40:11",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "futureframeai",
"github_project": "futureframe",
"travis_ci": false,
"coveralls": false,
"github_actions": true,
"lcname": "futureframe"
}
JSON
