ai-dan-core
High-performance neural network library with C++ backend and Python bindings.
Features
- Fast C++ backend with adaptive multi-threading
- Simple Python API for easy integration
- Built-in MNIST support for quick experimentation
- Flexible architecture supporting multiple activation functions and loss functions
- Model persistence - save and load trained models
Installation
pip install ai-dan-core
Quick Start
import ai_dan
# Create a neural network
network = ai_dan.Network([784, 256, 128, 10],
ai_dan.LEAKY_RELU,
ai_dan.SOFTMAX)
# Configure training
optimizer = ai_dan.Optimizer(learning_rate=0.001, batch_size=32)
loss = ai_dan.CrossEntropy()
# Train on MNIST
trainer = ai_dan.Trainer(network, optimizer, loss)
trainer.train(ai_dan.MNIST())
# Save the model
network.save("model.ai-dan")
# Load and benchmark
loaded = ai_dan.Network.load("model.ai-dan")
benchmarker = ai_dan.Benchmarker(loaded, ai_dan.MNISTBenchmark())
benchmarker.benchmark()
Activation Functions
ai_dan.SIGMOIDai_dan.RELUai_dan.LEAKY_RELUai_dan.TANHai_dan.SOFTMAX
Loss Functions
ai_dan.MSE(Mean Squared Error)ai_dan.CROSS_ENTROPY
Training with Callbacks
Monitor training progress with custom callbacks:
def on_step(step, loss_val, input_vec, label_vec, output_vec):
if step % 1000 == 0:
print(f"Step {step}, Loss: {loss_val:.4f}")
trainer = ai_dan.Trainer(network, optimizer, loss, on_step)
trainer.train(mnist_data)
Performance
- Training on MNIST: ~20 seconds for 60,000 samples
- Throughput: ~3,000 samples/second
- Typical accuracy: 96%+ on MNIST test set
Requirements
- Python 3.8+
- C++23 compatible compiler (GCC 13+, Clang 16+)
License
MIT
1 maintainer
Metadata
1 maintainer
Release files for ai-dan-core 0.1.4
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| ai_dan_core-0.1.4.tar.gz | 16.4 MB | Details |
Built distributions (wheels)
Total release size: 19.2 MB
Release files / ai_dan_core-0.1.4.tar.gz
| Download URL | ai_dan_core-0.1.4.tar.gz |
|---|---|
| Size | 16.4 MB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
d60fb225a07d3f77c063e5450656c0b29110209e08194f1ab9ad1f43d31020a7
|
|
BLAKE2b-256 checksum How to use checksums |
97b91bdf735158232ebb4e4d0745245c6e9da3720c86196fad70c73df6aaf6ed
|
| Upload date | |
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jan 4, 2026.
Transparency logRelease files / ai_dan_core-0.1.4-cp312-cp312-win_amd64.whl
| Download URL | ai_dan_core-0.1.4-cp312-cp312-win_amd64.whl |
|---|---|
| Size | 308.2 kB |
| Tags | CPython 3.12 Windows x86-64 |
|
SHA-256 checksum How to use checksums |
99fe3264d078b30ff54eb3406949e542e412f897c6d9b14752f59dacf13b235c
|
|
BLAKE2b-256 checksum How to use checksums |
fd58838b8d3fb0f488ffa31c7e3ab5ecdd6db641488439c68041726917005e21
|
| Upload date | |
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jan 4, 2026.
Transparency logRelease files / ai_dan_core-0.1.4-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl
| Download URL | ai_dan_core-0.1.4-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl |
|---|---|
| Size | 302.2 kB |
| Tags | CPython 3.12 Linux glibc 2.27+ x86-64 Linux glibc 2.28+ x86-64 |
|
SHA-256 checksum How to use checksums |
b3e662fe0ab71c671da5e1c7ee5e0ef64beb23727c99bbd114e379cbe8e9893c
|
|
BLAKE2b-256 checksum How to use checksums |
a86a80a0729c917113b963b953e4a74f54c5cda638c0166545e12e6b4cf8b755
|
| Upload date | |
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jan 4, 2026.
Transparency logRelease files / ai_dan_core-0.1.4-cp312-cp312-macosx_13_0_x86_64.whl
| Download URL | ai_dan_core-0.1.4-cp312-cp312-macosx_13_0_x86_64.whl |
|---|---|
| Size | 170.9 kB |
| Tags | CPython 3.12 macOS 13.0+ x86-64 |
|
SHA-256 checksum How to use checksums |
1acad22a45f36587e4a3426b38aece04f5a1d2fb75c267be1938ddba36f26d73
|
|
BLAKE2b-256 checksum How to use checksums |
fd5f19e63cedfc7f2fec54f66938613cb7a43f76e906ac087ddd2569f8f07f4a
|
| Upload date | |
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jan 4, 2026.
Transparency logRelease files / ai_dan_core-0.1.4-cp312-cp312-macosx_13_0_arm64.whl
| Download URL | ai_dan_core-0.1.4-cp312-cp312-macosx_13_0_arm64.whl |
|---|---|
| Size | 165.5 kB |
| Tags | CPython 3.12 macOS 13.0+ ARM64 |
|
SHA-256 checksum How to use checksums |
671fe15184f31feb4fa1655943830ac0c8314595cb472a55975c544a61e9f883
|
|
BLAKE2b-256 checksum How to use checksums |
5e20d4fabd7747989cc29202fda7af262fb014637ea8b4d301f35d648215fe54
|
| Upload date | |
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jan 4, 2026.
Transparency logRelease files / ai_dan_core-0.1.4-cp311-cp311-win_amd64.whl
| Download URL | ai_dan_core-0.1.4-cp311-cp311-win_amd64.whl |
|---|---|
| Size | 308.5 kB |
| Tags | CPython 3.11 Windows x86-64 |
|
SHA-256 checksum How to use checksums |
ea857e70f71f68028a7b0f1f700d6411e8b06f85b83c7f1f53ad24a93e16301a
|
|
BLAKE2b-256 checksum How to use checksums |
77d183d93aace1982a95c638f316bd5a72dffaa68ddb59b0c6f6239fbec5b291
|
| Upload date | |
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jan 4, 2026.
Transparency logRelease files / ai_dan_core-0.1.4-cp311-cp311-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl
| Download URL | ai_dan_core-0.1.4-cp311-cp311-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl |
|---|---|
| Size | 303.1 kB |
| Tags | CPython 3.11 Linux glibc 2.27+ x86-64 Linux glibc 2.28+ x86-64 |
|
SHA-256 checksum How to use checksums |
667bd47806da9ecdeded9fa8535c31754eab29464a1228175f148406ec26e8b7
|
|
BLAKE2b-256 checksum How to use checksums |
b444092ae8df18e6dfe736d88cc10744926d8f1a7b76f509fd23584a2578a004
|
| Upload date | |
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jan 4, 2026.
Transparency logRelease files / ai_dan_core-0.1.4-cp311-cp311-macosx_13_0_x86_64.whl
| Download URL | ai_dan_core-0.1.4-cp311-cp311-macosx_13_0_x86_64.whl |
|---|---|
| Size | 171.1 kB |
| Tags | CPython 3.11 macOS 13.0+ x86-64 |
|
SHA-256 checksum How to use checksums |
83ba29635f8f3b63309b1bba750924300eef94616b5f9a37f1121aaee640e45a
|
|
BLAKE2b-256 checksum How to use checksums |
b8aad54e87528fb5074cdf14019a543de82b724b066a64a3a628f47f6e36943b
|
| Upload date | |
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jan 4, 2026.
Transparency logRelease files / ai_dan_core-0.1.4-cp311-cp311-macosx_13_0_arm64.whl
| Download URL | ai_dan_core-0.1.4-cp311-cp311-macosx_13_0_arm64.whl |
|---|---|
| Size | 165.6 kB |
| Tags | CPython 3.11 macOS 13.0+ ARM64 |
|
SHA-256 checksum How to use checksums |
e82729aa863fc8b0c40d1b188c3e57e0c532ddcf176c9edce241a4efad36225d
|
|
BLAKE2b-256 checksum How to use checksums |
ea3f16c60459a2234c0938958e32f517ee2621f3422413ec92c3798f4f941bcc
|
| Upload date | |
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jan 4, 2026.
Transparency logRelease files / ai_dan_core-0.1.4-cp310-cp310-win_amd64.whl
| Download URL | ai_dan_core-0.1.4-cp310-cp310-win_amd64.whl |
|---|---|
| Size | 307.0 kB |
| Tags | CPython 3.10 Windows x86-64 |
|
SHA-256 checksum How to use checksums |
f939c47cdc7f9d31604b207fc2170cc6d8ed53cfd96d728b16e740d4e18905b7
|
|
BLAKE2b-256 checksum How to use checksums |
7982099e28ba31c17262e549bcc9e16c082dca5e79f687061829c23eb22ade30
|
| Upload date | |
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jan 4, 2026.
Transparency logRelease files / ai_dan_core-0.1.4-cp310-cp310-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl
| Download URL | ai_dan_core-0.1.4-cp310-cp310-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl |
|---|---|
| Size | 301.9 kB |
| Tags | CPython 3.10 Linux glibc 2.27+ x86-64 Linux glibc 2.28+ x86-64 |
|
SHA-256 checksum How to use checksums |
f8668efa399e82a1a84678aac64e3311090b9f3190d2aa0d426f09a47e45b749
|
|
BLAKE2b-256 checksum How to use checksums |
8c72334d47d94d4ae5273c9f48895255a5252c46d8241d2b37ea759457da9476
|
| Upload date | |
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jan 4, 2026.
Transparency logRelease files / ai_dan_core-0.1.4-cp310-cp310-macosx_13_0_x86_64.whl
| Download URL | ai_dan_core-0.1.4-cp310-cp310-macosx_13_0_x86_64.whl |
|---|---|
| Size | 169.7 kB |
| Tags | CPython 3.10 macOS 13.0+ x86-64 |
|
SHA-256 checksum How to use checksums |
f47c193af950b2989f14987278d9cdd99de21c89c4f61022ca1216d44db0d21c
|
|
BLAKE2b-256 checksum How to use checksums |
89f0e6916aa0f43d0c1f33fd62f660096196cd7c84d891e9ada1d96554157059
|
| Upload date | |
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jan 4, 2026.
Transparency logRelease files / ai_dan_core-0.1.4-cp310-cp310-macosx_13_0_arm64.whl
| Download URL | ai_dan_core-0.1.4-cp310-cp310-macosx_13_0_arm64.whl |
|---|---|
| Size | 164.3 kB |
| Tags | CPython 3.10 macOS 13.0+ ARM64 |
|
SHA-256 checksum How to use checksums |
d0284123909d93fe63bb6c14cbf61b7fa46d7336af66db08ff2d25d0674989a3
|
|
BLAKE2b-256 checksum How to use checksums |
8986ace4ab1ecc973f5bf42e586e5d396d76337b85396865db6bca719e9995e1
|
| Upload date | |
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jan 4, 2026.
Transparency log
