[](https://pypi.org/project/litten/)     
<img src="images/bg.png" width="1000">
# Litten
Litten is a python package to visualize sequential Tensorflow(keras) neural network model architectures, Get layers summary in details, visualize conv layers filters and featurmaps.
## Installation
To install latest version from PyPi
```bash
$ pip install litten
```
## How to use it?
1. Create your neural network archtictures
```python
model = keras.models.Sequential(
[
keras.layers.Conv2D(filters = 32, kernel_size = 3, input_shape = [150, 150, 3]),
...
keras.layers.Dense(units = 2, activation = 'softmax')
]
)
# or using functional API
input = keras.Input(shape=(28, 28, 1), name="img")
...
output = keras.layers.Dense(units = 2, activation = 'softmax')(x)
model = keras.Model(input, output, name="Model")
```
2. Import LayersSummary, ModelVisualizer and create objects
```python
from litten import LayersSummary, ModelVisualizer
```
3. To get Layers Summaries
```python
summary = LayersSummary()
summary.show_layers_summaries(model)
```
Output:
```plaintext
=================================================================================================================
Layer 1: InputLayer | Attributes
----------------------------------------
built : True
sparse : False
ragged : False
batch_size : None
is_placeholder : True
=================================================================================================================
Layer 2: Conv2D | Attributes
----------------------------------------
rank : 2
filters : 16
groups : 1
kernel_size : (3, 3)
strides : (1, 1)
padding : valid
data_format : channels_last
dilation_rate : (1, 1)
activation : <function relu at 0x7f6b7038fee0>
use_bias : True
kernel_initializer : <keras.initializers.initializers_v2.GlorotUniform object at 0x7f6aee189610>
bias_initializer : <keras.initializers.initializers_v2.Zeros object at 0x7f6aee189af0>
kernel_regularizer : None
bias_regularizer : None
kernel_constraint : None
bias_constraint : None
built : True
=================================================================================================================
....
```
4. To visualize model architecture
```Plaintext
ModelVisualizer.visualize_model(
show_names=False,
show_properties=False,
show_connectors=False,
pallete='default'
)
```
Example 1
```python
vis = ModelVisualizer(model)
vis.visualize_model()
```
<img src="images/basic.png" width="1000">
Example 2
```python
vis.visualize_model(show_names=True)
```
<img src="images/names.png" width="1000">
Example 3
```python
vis.visualize_model(show_names=True, show_connectors = True)
```
<img src="images/connectors.png" width="1000">
Exampel 4
```python
vis.visualize_model(show_names=True, show_connectors = True, show_properties=True)
```
<img src="images/properties.png" width="1000" height="">
Example 5
You can choose one from these palettes `default`, `red`, `green`, `blue`, `yellow`, `brown`, `purple`, `gray`
```python
vis.visualize_model(show_names=True, show_connectors=True, palette=<palette>)
```
<img src="images/gray.png" width="600" align="center">
<img src="images/blue.png" width="600" align="center">
<img src="images/red.png" width="600" align="center">
<img src="images/brown.png" width="600" align="center">
<img src="images/purple.png" width="600" align="center">
5. To visualize Conv filters
```plaintext
ModelVisualizer.visualize_filters(
cmap = 'gray' # matplotlib cmaps
)
```
Example
```python
vis = ModelVisualizer(model)
vis.visualize_filters('Blues')
```
<img src="images/filters.png" width="500">
6. To visualize Features map
```plaintext
ModelVisualizer.visualize_featuremap(
input_image,
cmap = 'gray' # matplotlib cmaps
)
```
Example
```python
vis = ModelVisualizer(model)
vis.visualize_featuremap(input_image, 'Blues')
```
<img src="images/feature_map1.png" width="500">
<img src="images/feature_map2.png" width="1000">
<img src="images/feature_map3.png" width="1000">
<img src="images/feature_map4.png" width="1000">
## Contributing to `litten`
To contribute to litten, follow these steps:
1. Fork this repository.
2. Create a branch: `git checkout -b <branch_name>`.
3. Make your changes and commit them: `git commit -m '<commit_message>'`
4. Push to the original branch: `git push origin <project_name>/<location>`
5. Create the pull request.
Alternatively see the GitHub documentation on [creating a pull request](https://help.github.com/en/github/collaborating-with-issues-and-pull-requests/creating-a-pull-request).
## License
Open source licensed under the MIT license (see _LICENSE_ file for details).
## Tools
- Python
- PIL
- Matplotlib
- pytest
Raw data
{
"_id": null,
"home_page": "",
"name": "litten",
"maintainer": "",
"docs_url": null,
"requires_python": ">=3.6",
"maintainer_email": "",
"keywords": "python,tensorflow,keras,PIL,pillow,visualize models,visualize",
"author": "Hossam Asaad",
"author_email": "hossamasaad10@gmail.com",
"download_url": "https://files.pythonhosted.org/packages/ee/52/f747eb5c982120bb1630a22e7dada4ec720ca38f61abb62f5931a168b177/litten-0.1.0.tar.gz",
"platform": null,
"description": "\n[](https://pypi.org/project/litten/)     \n\n<img src=\"images/bg.png\" width=\"1000\">\n\n# Litten\nLitten is a python package to visualize sequential Tensorflow(keras) neural network model architectures, Get layers summary in details, visualize conv layers filters and featurmaps. \n\n## Installation\nTo install latest version from PyPi\n```bash\n$ pip install litten\n```\n\n## How to use it?\n\n1. Create your neural network archtictures\n\n```python\nmodel = keras.models.Sequential(\n [\n keras.layers.Conv2D(filters = 32, kernel_size = 3, input_shape = [150, 150, 3]),\n ...\n keras.layers.Dense(units = 2, activation = 'softmax')\n ]\n)\n\n# or using functional API\n\ninput = keras.Input(shape=(28, 28, 1), name=\"img\")\n...\noutput = keras.layers.Dense(units = 2, activation = 'softmax')(x)\nmodel = keras.Model(input, output, name=\"Model\")\n```\n\n2. Import LayersSummary, ModelVisualizer and create objects\n\n```python\nfrom litten import LayersSummary, ModelVisualizer\n```\n\n3. To get Layers Summaries\n```python\nsummary = LayersSummary()\nsummary.show_layers_summaries(model)\n```\nOutput:\n```plaintext\n=================================================================================================================\nLayer 1: InputLayer | Attributes\n----------------------------------------\nbuilt : True\nsparse : False\nragged : False\nbatch_size : None\nis_placeholder : True\n=================================================================================================================\nLayer 2: Conv2D | Attributes\n----------------------------------------\nrank : 2\nfilters : 16\ngroups : 1\nkernel_size : (3, 3)\nstrides : (1, 1)\npadding : valid\ndata_format : channels_last\ndilation_rate : (1, 1)\nactivation : <function relu at 0x7f6b7038fee0>\nuse_bias : True\nkernel_initializer : <keras.initializers.initializers_v2.GlorotUniform object at 0x7f6aee189610>\nbias_initializer : <keras.initializers.initializers_v2.Zeros object at 0x7f6aee189af0>\nkernel_regularizer : None\nbias_regularizer : None\nkernel_constraint : None\nbias_constraint : None\nbuilt : True\n=================================================================================================================\n....\n```\n\n4. To visualize model architecture\n ```Plaintext\n ModelVisualizer.visualize_model(\n show_names=False,\n show_properties=False,\n show_connectors=False,\n pallete='default'\n )\n ```\n Example 1\n ```python\n vis = ModelVisualizer(model)\n vis.visualize_model()\n ```\n <img src=\"images/basic.png\" width=\"1000\">\n \n \n Example 2\n ```python\n vis.visualize_model(show_names=True)\n ```\n <img src=\"images/names.png\" width=\"1000\">\n\n Example 3\n ```python\n vis.visualize_model(show_names=True, show_connectors = True)\n ```\n <img src=\"images/connectors.png\" width=\"1000\">\n\n Exampel 4\n ```python\n vis.visualize_model(show_names=True, show_connectors = True, show_properties=True)\n ```\n <img src=\"images/properties.png\" width=\"1000\" height=\"\">\n\n Example 5\n You can choose one from these palettes `default`, `red`, `green`, `blue`, `yellow`, `brown`, `purple`, `gray`\n ```python\n vis.visualize_model(show_names=True, show_connectors=True, palette=<palette>)\n ```\n <img src=\"images/gray.png\" width=\"600\" align=\"center\">\n <img src=\"images/blue.png\" width=\"600\" align=\"center\">\n <img src=\"images/red.png\" width=\"600\" align=\"center\">\n <img src=\"images/brown.png\" width=\"600\" align=\"center\">\n <img src=\"images/purple.png\" width=\"600\" align=\"center\">\n\n\n5. To visualize Conv filters\n ```plaintext\n ModelVisualizer.visualize_filters(\n cmap = 'gray' # matplotlib cmaps\n )\n ```\n Example\n ```python\n vis = ModelVisualizer(model)\n vis.visualize_filters('Blues')\n ```\n <img src=\"images/filters.png\" width=\"500\">\n\n6. To visualize Features map\n ```plaintext\n ModelVisualizer.visualize_featuremap(\n input_image,\n cmap = 'gray' # matplotlib cmaps\n )\n ```\n Example\n ```python\n vis = ModelVisualizer(model)\n vis.visualize_featuremap(input_image, 'Blues')\n ```\n <img src=\"images/feature_map1.png\" width=\"500\">\n <img src=\"images/feature_map2.png\" width=\"1000\">\n <img src=\"images/feature_map3.png\" width=\"1000\">\n <img src=\"images/feature_map4.png\" width=\"1000\">\n\n\n## Contributing to `litten`\n\nTo contribute to litten, follow these steps:\n\n1. Fork this repository.\n2. Create a branch: `git checkout -b <branch_name>`.\n3. Make your changes and commit them: `git commit -m '<commit_message>'`\n4. Push to the original branch: `git push origin <project_name>/<location>`\n5. Create the pull request.\n\nAlternatively see the GitHub documentation on [creating a pull request](https://help.github.com/en/github/collaborating-with-issues-and-pull-requests/creating-a-pull-request).\n\n## License\nOpen source licensed under the MIT license (see _LICENSE_ file for details).\n\n## Tools\n- Python\n- PIL\n- Matplotlib\n- pytest\n\n",
"bugtrack_url": null,
"license": "",
"summary": "A python package to visualize sequentail tensorflow model archtictures",
"version": "0.1.0",
"split_keywords": [
"python",
"tensorflow",
"keras",
"pil",
"pillow",
"visualize models",
"visualize"
],
"urls": [
{
"comment_text": "",
"digests": {
"md5": "32007b1a7a4d757e570459de4ff64332",
"sha256": "aaffd13a7444d4df44b52317f018e7d6bb397b1f8ce0b0c26cd515dc3c0f59d3"
},
"downloads": -1,
"filename": "litten-0.1.0-py3-none-any.whl",
"has_sig": false,
"md5_digest": "32007b1a7a4d757e570459de4ff64332",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.6",
"size": 12042,
"upload_time": "2022-12-23T13:57:26",
"upload_time_iso_8601": "2022-12-23T13:57:26.907459Z",
"url": "https://files.pythonhosted.org/packages/f5/59/ef367da99633b72c060911ae78ba3c3156f583b5af005209b0b2dd269b8b/litten-0.1.0-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"md5": "de1e75cf9439aeab7ca58bcbda643f32",
"sha256": "7d3135de7c89ee9dfb901ed18cf82835b88d6fc24a0856076f21ee85be36b323"
},
"downloads": -1,
"filename": "litten-0.1.0.tar.gz",
"has_sig": false,
"md5_digest": "de1e75cf9439aeab7ca58bcbda643f32",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.6",
"size": 12385,
"upload_time": "2022-12-23T13:57:28",
"upload_time_iso_8601": "2022-12-23T13:57:28.952888Z",
"url": "https://files.pythonhosted.org/packages/ee/52/f747eb5c982120bb1630a22e7dada4ec720ca38f61abb62f5931a168b177/litten-0.1.0.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2022-12-23 13:57:28",
"github": false,
"gitlab": false,
"bitbucket": false,
"lcname": "litten"
}
JSON