AIronTools (Beta) is a Python library that provides the user with deep learning tools built to work with tensorflow as a backend.
Project description
AIronTools
AIronTools (Beta) is a Python library that provides the user with higher level state-of-the-art deep learning tools built to work with tensorflow as a backend. The main goal of this repository is to enable fast model design for both POCs and production.
Key features:
- Out-of-the-box models ready to be used.
- Block constructor to build customised blocks/models.
- Layer constructor to build customised layers such as sequential, convolutional, self-attention or dense, and combinations of them.
- Preprocessing tools.
- On the fly non-topological hyper-parameter optimization. For now only the dropout regularization is compatible with this feature, in the future others such as l1 and l2 regularization will be compatible too.
- Latent representations for visualization purposes.
Installation
pip install airontools
Custom Keras subclass to build a variational autoencoder (VAE) with airontools and compatible with aironsuit
import numpy as np
import tensorflow as tf
from airontools.constructors.models.unsupervised.vae import VAE
from numpy.random import normal
tabular_data = np.concatenate(
[
normal(loc=0.5, scale=1, size=(100, 10)),
normal(loc=-0.5, scale=1, size=(100, 10)),
]
)
model = VAE(
input_shape=tabular_data.shape[1:],
latent_dim=3,
)
model.compile(optimizer=tf.keras.optimizers.Adam(learning_rate=0.001))
model.fit(
tabular_data,
epochs=10,
)
print("VAE evaluation:", float(model.evaluate(tabular_data)["loss"]))
More examples
see usage examples in aironsuit/examples
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
airontools-0.1.25.tar.gz
(14.6 kB
view hashes)
Built Distribution
Close
Hashes for airontools-0.1.25-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 0ab334fd8f512e808777b46748cd9a0f07a33b34c8239534d1cb18c921014a98 |
|
MD5 | 22fa0abd903700226e4443bd6eaa4f10 |
|
BLAKE2b-256 | 68a30a8457a502f2b10d67e332788880b07e93e2bba6310d6047ef6b314de6e2 |