tensorflow-wavelets is an implementation of Custom Layers for Neural Networks:
- *Discrete Wavelets Transform Layer*
- *Duel Tree Complex Wavelets Transform Layer*
- *Multi Wavelets Transform Layer*
#
```
git clone https://github.com/Timorleiderman/tensorflow-wavelets.git
cd tensorflow-wavelets
pip install -r requirements.txt
```
## Installation
#### tested with python 3.8
```
pip install tensorflow-wavelets
```
# Usage
```
from tensorflow import keras
import tensorflow_wavelets.Layers.DWT as DWT
import tensorflow_wavelets.Layers.DTCWT as DTCWT
import tensorflow_wavelets.Layers.DMWT as DMWT
# Custom Activation function Layer
import tensorflow_wavelets.Layers.Threshold as Threshold
```
# Examples
## DWT(name="haar", concat=0)
### "name" can be found in pywt.wavelist(family)
### concat = 0 means to split to 4 smaller layers
```
from tensorflow import keras
model = keras.Sequential()
model.add(keras.Input(shape=(28, 28, 1)))
model.add(DWT.DWT(name="haar",concat=0))
model.add(keras.layers.Flatten())
model.add(keras.layers.Dense(nb_classes, activation="softmax"))
model.summary()
```
_________________________________________________________________
Layer (type) Output Shape Param #
=================================================================
dwt_9_haar (DWT) (None, 14, 14, 4) 0
_________________________________________________________________
flatten_9 (Flatten) (None, 784) 0
_________________________________________________________________
dense_9 (Dense) (None, 10) 7850
=================================================================
Total params: 7,850
Trainable params: 7,850
Non-trainable params: 0
_________________________________________________________________
### name = "db4" concat = 1
```
model = keras.Sequential()
model.add(keras.layers.InputLayer(input_shape=(28, 28, 1)))
model.add(DWT.DWT(name="db4", concat=1))
model.summary()
```
Model: "sequential"
_________________________________________________________________
Layer (type) Output Shape Param #
=================================================================
dwt_db4 (DWT) (None, 34, 34, 1) 0
=================================================================
Total params: 0
Trainable params: 0
Non-trainable params: 0
_________________________________________________________________
# DMWT
### functional example with Sure Threshold
```
x_inp = keras.layers.Input(shape=(512, 512, 1))
x = DMWT.DMWT("ghm")(x_inp)
x = Threshold.Threshold(algo='sure', mode='hard')(x) # use "soft" or "hard"
x = DMWT.IDMWT("ghm")(x)
model = keras.models.Model(x_inp, x, name="MyModel")
model.summary()
```
Model: "MyModel"
_________________________________________________________________
Layer (type) Output Shape Param #
=================================================================
input_1 (InputLayer) [(None, 512, 512, 1)] 0
_________________________________________________________________
dmwt (DMWT) (None, 1024, 1024, 1) 0
_________________________________________________________________
sure_threshold (SureThreshol (None, 1024, 1024, 1) 0
_________________________________________________________________
idmwt (IDMWT) (None, 512, 512, 1) 0
=================================================================
Total params: 0
Trainable params: 0
Non-trainable params: 0
_________________________________________________________________
## PyPi upload:
```
pip install --upgrade build
pip install --upgrade twine
python -m build
python -m twine upload --repository pypi dist/*
```
If our open source codes are helpful for your research, please cite our
[technical report:](https://www.mdpi.com/1099-4300/26/10/836)
```
@Article{e26100836,
AUTHOR = {Leiderman, Timor and Ben Ezra, Yosef},
TITLE = {Information Bottleneck Driven Deep Video Compression—IBOpenDVCW},
JOURNAL = {Entropy},
VOLUME = {26},
YEAR = {2024},
NUMBER = {10},
ARTICLE-NUMBER = {836},
URL = {https://www.mdpi.com/1099-4300/26/10/836},
ISSN = {1099-4300},
DOI = {10.3390/e26100836}
}
```
**Free Software, Hell Yeah!**
Raw data
{
"_id": null,
"home_page": "https://github.com/Timorleiderman/tensorflow-wavelets",
"name": "tensorflow-wavelets",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.6",
"maintainer_email": null,
"keywords": "Wavelets, Tensorflow",
"author": "Timor Leiderman",
"author_email": "Timorleiderman@gmail.com",
"download_url": "https://files.pythonhosted.org/packages/b5/42/eefffee3ea40b517cee480c4243371f102c647f6355f658d75d670af6795/tensorflow_wavelets-1.0.31.tar.gz",
"platform": null,
"description": "tensorflow-wavelets is an implementation of Custom Layers for Neural Networks:\n- *Discrete Wavelets Transform Layer*\n- *Duel Tree Complex Wavelets Transform Layer*\n- *Multi Wavelets Transform Layer*\n\n\n\n# \n```\ngit clone https://github.com/Timorleiderman/tensorflow-wavelets.git\ncd tensorflow-wavelets\npip install -r requirements.txt\n```\n## Installation\n#### tested with python 3.8\n```\npip install tensorflow-wavelets\n```\n# Usage\n```\nfrom tensorflow import keras\nimport tensorflow_wavelets.Layers.DWT as DWT\nimport tensorflow_wavelets.Layers.DTCWT as DTCWT\nimport tensorflow_wavelets.Layers.DMWT as DMWT\n\n# Custom Activation function Layer\nimport tensorflow_wavelets.Layers.Threshold as Threshold\n```\n\n# Examples\n## DWT(name=\"haar\", concat=0)\n### \"name\" can be found in pywt.wavelist(family)\n### concat = 0 means to split to 4 smaller layers\n\n```\nfrom tensorflow import keras\nmodel = keras.Sequential()\nmodel.add(keras.Input(shape=(28, 28, 1)))\nmodel.add(DWT.DWT(name=\"haar\",concat=0))\nmodel.add(keras.layers.Flatten())\nmodel.add(keras.layers.Dense(nb_classes, activation=\"softmax\"))\nmodel.summary()\n```\n\n _________________________________________________________________\n Layer (type) Output Shape Param #\n =================================================================\n dwt_9_haar (DWT) (None, 14, 14, 4) 0\n _________________________________________________________________\n flatten_9 (Flatten) (None, 784) 0\n _________________________________________________________________\n dense_9 (Dense) (None, 10) 7850\n =================================================================\n Total params: 7,850\n Trainable params: 7,850\n Non-trainable params: 0\n _________________________________________________________________\n\n### name = \"db4\" concat = 1\n```\n\nmodel = keras.Sequential()\nmodel.add(keras.layers.InputLayer(input_shape=(28, 28, 1)))\nmodel.add(DWT.DWT(name=\"db4\", concat=1))\nmodel.summary()\n```\n\n Model: \"sequential\"\n _________________________________________________________________\n Layer (type) Output Shape Param #\n =================================================================\n dwt_db4 (DWT) (None, 34, 34, 1) 0\n =================================================================\n Total params: 0\n Trainable params: 0\n Non-trainable params: 0\n _________________________________________________________________\n\n# DMWT\n### functional example with Sure Threshold\n```\n\nx_inp = keras.layers.Input(shape=(512, 512, 1))\nx = DMWT.DMWT(\"ghm\")(x_inp)\nx = Threshold.Threshold(algo='sure', mode='hard')(x) # use \"soft\" or \"hard\"\nx = DMWT.IDMWT(\"ghm\")(x)\nmodel = keras.models.Model(x_inp, x, name=\"MyModel\")\nmodel.summary()\n```\n Model: \"MyModel\"\n _________________________________________________________________\n Layer (type) Output Shape Param #\n =================================================================\n input_1 (InputLayer) [(None, 512, 512, 1)] 0\n _________________________________________________________________\n dmwt (DMWT) (None, 1024, 1024, 1) 0\n _________________________________________________________________\n sure_threshold (SureThreshol (None, 1024, 1024, 1) 0\n _________________________________________________________________\n idmwt (IDMWT) (None, 512, 512, 1) 0\n =================================================================\n Total params: 0\n Trainable params: 0\n Non-trainable params: 0\n _________________________________________________________________\n\n\n## PyPi upload:\n```\npip install --upgrade build\npip install --upgrade twine\npython -m build\npython -m twine upload --repository pypi dist/*\n\n```\n\nIf our open source codes are helpful for your research, please cite our\n[technical report:](https://www.mdpi.com/1099-4300/26/10/836)\n```\n@Article{e26100836,\nAUTHOR = {Leiderman, Timor and Ben Ezra, Yosef},\nTITLE = {Information Bottleneck Driven Deep Video Compression\u2014IBOpenDVCW},\nJOURNAL = {Entropy},\nVOLUME = {26},\nYEAR = {2024},\nNUMBER = {10},\nARTICLE-NUMBER = {836},\nURL = {https://www.mdpi.com/1099-4300/26/10/836},\nISSN = {1099-4300},\nDOI = {10.3390/e26100836}\n}\n```\n\n**Free Software, Hell Yeah!**\n",
"bugtrack_url": null,
"license": null,
"summary": "Tensorflow wavelet Layers",
"version": "1.0.31",
"project_urls": {
"Bug Tracker": "https://github.com/Timorleiderman/tensorflow-wavelets/issues",
"Homepage": "https://github.com/Timorleiderman/tensorflow-wavelets"
},
"split_keywords": [
"wavelets",
" tensorflow"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "d98e5d3f81de78a82d71a0f71f178731c04152fc2df7a7ca4d0c568e5118acb4",
"md5": "decfbee1eeb433cadcb6100e552a7348",
"sha256": "3c08f33ebfb92da409f2fd02a7c003734e2aa6cf83d3451c5a3c18d2d00cc2e0"
},
"downloads": -1,
"filename": "tensorflow_wavelets-1.0.31-py3-none-any.whl",
"has_sig": false,
"md5_digest": "decfbee1eeb433cadcb6100e552a7348",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.6",
"size": 26071,
"upload_time": "2024-12-20T11:20:00",
"upload_time_iso_8601": "2024-12-20T11:20:00.583640Z",
"url": "https://files.pythonhosted.org/packages/d9/8e/5d3f81de78a82d71a0f71f178731c04152fc2df7a7ca4d0c568e5118acb4/tensorflow_wavelets-1.0.31-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "b542eefffee3ea40b517cee480c4243371f102c647f6355f658d75d670af6795",
"md5": "0d45d6588bae8551faa3838dcde8c4d8",
"sha256": "475940a0ce5821f7e0a380d7fecb7d9e10e34f5f542098b2804c39cc5ec40f5c"
},
"downloads": -1,
"filename": "tensorflow_wavelets-1.0.31.tar.gz",
"has_sig": false,
"md5_digest": "0d45d6588bae8551faa3838dcde8c4d8",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.6",
"size": 21880,
"upload_time": "2024-12-20T11:20:02",
"upload_time_iso_8601": "2024-12-20T11:20:02.911958Z",
"url": "https://files.pythonhosted.org/packages/b5/42/eefffee3ea40b517cee480c4243371f102c647f6355f658d75d670af6795/tensorflow_wavelets-1.0.31.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-12-20 11:20:02",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "Timorleiderman",
"github_project": "tensorflow-wavelets",
"travis_ci": false,
"coveralls": false,
"github_actions": true,
"requirements": [
{
"name": "scipy",
"specs": [
[
"==",
"1.5.2"
]
]
},
{
"name": "matplotlib",
"specs": [
[
"~=",
"3.4.3"
]
]
},
{
"name": "tensorflow",
"specs": [
[
"==",
"2.6.1"
]
]
},
{
"name": "keras",
"specs": [
[
"==",
"2.6"
]
]
},
{
"name": "psnr-hvsm",
"specs": [
[
"==",
"0.1.0"
]
]
},
{
"name": "opencv-python",
"specs": []
},
{
"name": "scikit-image",
"specs": []
},
{
"name": "pywavelets",
"specs": []
},
{
"name": "tensorflow_probability",
"specs": [
[
"==",
"0.14.1"
]
]
},
{
"name": "numpy",
"specs": [
[
"==",
"1.19.2"
]
]
},
{
"name": "ipykernel",
"specs": []
},
{
"name": "protobuf",
"specs": [
[
"==",
"3.20.3"
]
]
},
{
"name": "pandas",
"specs": [
[
"==",
"1.4"
]
]
}
],
"lcname": "tensorflow-wavelets"
}
JSON
