Ensemble TabPFN
TabPFN is a transformer architecture prosposed by Hollman et al for classification on small tabular datasets. It is a Prior-Data Fitted Network that has been trained once and does not require fine tuning for new datasets. It works by approximating the distribution of new data to the prior synthetic data it has seen during training. In a machine learning pipeline, this network can be "fit" on a training dataset in under a second and can generate predictions for the test set in a single forward pass in the network. However there are limitations in the current architecture, namely, the training dataset can contain only upto 1000 inputs with upto 100 numerical features. In addition, the network can predict only upto 10 classes in a multi-class classification problem. With EnsembleTabPFN, we address two of these issues where we have extended the original model to work with datasets containing more than 1000 samples and 100 features. EnsembleTabPFN is fully compatible with Scikit-learn API and can be used in a modelling pipeline.
Installation
From source
git clone https://github.com/ersilia-os/ensemble-tabpfn.git
cd ensemble-tabpfn
pip install .
From PyPI
pip install ensemble-tabpfn
Using Poetry
git clone https://github.com/ersilia-os/ensemble-tabpfn.git
cd ensemble-tabpfn
poetry install --without dev,test,docs
Usage
from ensemble_tabpfn import EnsembleTabPFN
from sklearn.metrics import accuracy_score
clf = EnsembleTabPFN()
clf.fit(X_train, y_train)
y_hat = clf.predict(y_test)
acc = accuracy_score(y_test, y_hat)
Metadata
Release files for ensemble-tabpfn 0.1.1
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| ensemble_tabpfn-0.1.1.tar.gz | 17.9 kB | Without provenance | Details for ensemble_tabpfn-0.1.1.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| ensemble_tabpfn-0.1.1-py3-none-any.whl | Python 3 | none | any | Details for ensemble_tabpfn-0.1.1-py3-none-any.whl |
Total release size: 37.3 kB
Release files / ensemble_tabpfn-0.1.1.tar.gz
| Download URL | ensemble_tabpfn-0.1.1.tar.gz |
|---|---|
| Size | 17.9 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
a94941b44812430d49d7aa190d4d61f50bf5ec65c09b1db699c3440f2f1cbff4
|
|
BLAKE2b-256 checksum How to use checksums |
34d90fa0e03e134296fbbfd9c880dd02187df725d55fa8ec428652f9ae2cb28a
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
poetry/1.3.2 CPython/3.7.14 Linux/5.19.0-32-generic
|
Release files / ensemble_tabpfn-0.1.1-py3-none-any.whl
| Download URL | ensemble_tabpfn-0.1.1-py3-none-any.whl |
|---|---|
| Size | 19.4 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
7d9abdc544bf86974dae8aa8aa1adaa44ed847254c051ba3890fbca7d4863641
|
|
BLAKE2b-256 checksum How to use checksums |
9768f244f0654bd8f82d84dd970874e91fb7019281d87d5c229d4375a45bf89f
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
poetry/1.3.2 CPython/3.7.14 Linux/5.19.0-32-generic
|
Security for ensemble-tabpfn 0.1.1
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.