Keras based model builder
Project description
Alquimodelia
Alquimodelia is a Python package that provides a Keras-based forecast model builder.
It provides the arquitectures for CNN, LSTM, and Encoder Decoder, and even from imagery UNET. Any suggestions and tips are welcome. Use this to fastly have your forecast models ready to use!
Usage
To use Alquimodelia, follow these steps:
pip install alquimodelia
Since Alquimodelia is based on keras-core you can choose which backend to use, otherwise it will default to tensorflow.
To change backend change the KERAS-BACKEND
enviromental variable. Follow this.
To get an arquiteture you only need to have a simple configuration and call the module:
import alquimodelia
# The input arguments
input_args = {
"X_timeseries": 168,
"Y_timeseries": 24,
"n_features_train": 18,
"n_features_predict": 1,
}
# This is make a model with shapes:
# input_shape = (N, 168, 18)
# output_shape = (N, 24, 1)
forearch = alquimodelia.CNNArch(**input_args)
# Now for Vanilla and Stacked CNN:
architecture_args = {}
VanillaCNN = forearch.architecture(**architecture_args)
architecture_args = {"block_repetition": 2}
StackedCNN = forearch.architecture(**architecture_args)
# Keras Models ready to use:
VanillaCNN.summary()
StackedCNN.summary()
Contribution
Contributions to Alquimodelia are welcome! If you find any issues or have suggestions for improvement, please feel free to contribute. Make sure to update tests as appropriate and follow the contribution guidelines.
License
Alquimodelia is licensed under the MIT License, which allows you to use, modify, and distribute the package according to the terms of the license. For more details, please refer to the LICENSE file.
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
File details
Details for the file alquimodelia-0.0.6.tar.gz
.
File metadata
- Download URL: alquimodelia-0.0.6.tar.gz
- Upload date:
- Size: 28.5 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: poetry/1.8.4 CPython/3.11.0 Linux/6.5.0-1025-azure
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | 5b51779e487409f17e7a2bfabe515938fcdce87de23875158181eae848b69c6f |
|
MD5 | cddeb1833efcf6182069ffc606cc355d |
|
BLAKE2b-256 | d5c27e385f54970ca8e23a6a248414bbeaca802c9f3a0b3309d57fb20a44c0a8 |
Provenance
File details
Details for the file alquimodelia-0.0.6-py3-none-any.whl
.
File metadata
- Download URL: alquimodelia-0.0.6-py3-none-any.whl
- Upload date:
- Size: 34.1 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: poetry/1.8.4 CPython/3.11.0 Linux/6.5.0-1025-azure
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | 506908c366d5bd7735f9d23566e2cfa9f0f09dd6c4ac01ca40bd99f693b5ea02 |
|
MD5 | 2732a8d0c1e74b5841c36d5d3d955737 |
|
BLAKE2b-256 | c76e87ddc135e76464f2fbe376771b97b3dba6963ccd913beae7623e029c927c |