OptiML

Train neural networks via global mathematical optimisation (MINLP) instead of gradient descent.

OptiML translates a neural network architecture into a system of constraints and decision variables, then uses a MINLP solver (e.g. Couenne, SCIP) to find the weights that provably minimise the training loss. After solving, the model can be exported to PyTorch for inference.

Installation

pip install -e .            # core (numpy + pyomo)
pip install -e ".[pytorch]" # with PyTorch export support
pip install -e ".[all]"     # everything

You also need a MINLP solver accessible to Pyomo, for example Couenne or SCIP.

Quick start

import optiml
from optiml.losses import MSELoss

# Ultra-small Edge AI classifier — only 9 parameters
model = optiml.Sequential(
    optiml.Linear(2, 2),
    optiml.ReLU(M=10),
    optiml.Linear(2, 1),
)

model.fit(X_train, y_train, loss=MSELoss(reduction='sum'), solver='couenne')

# Export to PyTorch for inference / deployment
pytorch_model = model.export('pytorch')

Available layers

Layer	Description
`Linear(in, out)`	Fully-connected layer
`Conv1D(in_ch, out_ch, kernel)`	1-D convolution
`Conv2D(in_ch, out_ch, kernel)`	2-D convolution
`ReLU(M)`	ReLU via big-M formulation
`AvgPool2D(kernel)`	2-D average pooling
`Flatten()`	Flatten spatial dimensions

Available losses

Loss	Description
`MSELoss(reduction)`	Mean / sum of squared errors
`SSELoss()`	Sum of squared errors
`MAELoss(reduction)`	Mean / sum of absolute errors
`HuberLoss(delta)`	Smooth L1 loss

Export

After fitting, call model.export('pytorch') to get a torch.nn.Sequential with the optimal weights loaded.

Example

See examples/binary_classification.py for a full working example. It trains a 9-parameter Iris flower classifier (versicolor vs virginica from petal measurements) and compares OptiML with PyTorch + Adam:

OptiML finds the mathematically optimal weights in ~7 s, achieving 93.3 % test accuracy on 90 unseen samples.
PyTorch Adam has a 30 % failure rate (stuck at 50 % accuracy) and even the best restart (selected by training loss) only reaches 91.1 %.

Metadata

Release files for globoptiml 1.0.14

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for globoptiml 1.0.14
File	Size	Uploaded
globoptiml-1.0.14.tar.gz	20.0 kB	Mar 13, 2026	Details

Built distribution (wheel)

Table of built distributions (wheels) for globoptiml 1.0.14
File	Interpreter	ABI	Platform	Reset
globoptiml-1.0.14-py3-none-any.whl 21.9 kB Mar 13, 2026	Python 3	none	any	Details

Total release size: 41.9 kB

Release files / globoptiml-1.0.14.tar.gz

Download URL	globoptiml-1.0.14.tar.gz
Size	20.0 kB
Tags	Source
SHA-256 checksum How to use checksums	`dddb9fbe7a0fda49ef543ee0c3f32bee0c4fa976514fe985cc32f3294baa7106`
BLAKE2b-256 checksum How to use checksums	`7a0fad31a56afd56422a761b0f43e9ed8c7c2db458487d7976bb44babda990ab`
Upload date	Mar 13, 2026
Uploaded using Trusted Publishing? What is trusted publishing?	Yes
Uploaded via	`twine/6.1.0 CPython/3.13.7`

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.

Identity

Publishing platform: GitHub Actions
Publishing repository: github.com/modyf01/optiML
Publishing commit: github.com/modyf01/optiML/tree/77613ad54b186619f7d44de46b29ee5e9cae7073

Workflow

Publishing configuration: github.com/modyf01/optiML/blob/77613ad54b186619f7d44de46b29ee5e9cae7073/.github/workflows/publish_pypi.yml
Publishing logs: github.com/modyf01/optiML/actions/runs/23050703077/attempts/1

Artifact

Subject: globoptiml-1.0.14.tar.gz
SHA-256 checksum: dddb9fbe7a0fda49ef543ee0c3f32bee0c4fa976514fe985cc32f3294baa7106 Verifying attestations

Release files / globoptiml-1.0.14-py3-none-any.whl

Download URL	globoptiml-1.0.14-py3-none-any.whl
Size	21.9 kB
Tags	Python 3
SHA-256 checksum How to use checksums	`6e7aa7959cd2368a3b3cb699f36dbf02aaa8ccdc6a962ec3526592862c6243b2`
BLAKE2b-256 checksum How to use checksums	`cb97ffa6b64b4c7c8a16bbeedafa71f82245510ce75ac8cae4a3cac7d845e1a3`
Upload date	Mar 13, 2026
Uploaded using Trusted Publishing? What is trusted publishing?	Yes
Uploaded via	`twine/6.1.0 CPython/3.13.7`

Provenance

Identity

Publishing platform: GitHub Actions
Publishing repository: github.com/modyf01/optiML
Publishing commit: github.com/modyf01/optiML/tree/77613ad54b186619f7d44de46b29ee5e9cae7073

Workflow

Publishing configuration: github.com/modyf01/optiML/blob/77613ad54b186619f7d44de46b29ee5e9cae7073/.github/workflows/publish_pypi.yml
Publishing logs: github.com/modyf01/optiML/actions/runs/23050703077/attempts/1

Artifact

Subject: globoptiml-1.0.14-py3-none-any.whl
SHA-256 checksum: 6e7aa7959cd2368a3b3cb699f36dbf02aaa8ccdc6a962ec3526592862c6243b2 Verifying attestations