# RT Activation
A custom activation function for Keras/TensorFlow implementing Rectified Tangent Activation (RTA).
## Formula
f(x) = max(x, tanh(x))
## Installation
```bash
pip install rt-activation
```
## Usage
### Simple Usage (String-based)
```python
import keras
from keras import layers
import rt_activation # This registers the activation function
model = keras.Sequential([
keras.Input(shape=input_shape),
layers.Conv2D(32, kernel_size=(3, 3), activation="RTA"),
layers.MaxPooling2D(pool_size=(2, 2)),
layers.Conv2D(64, kernel_size=(3, 3), activation="RTA"),
layers.MaxPooling2D(pool_size=(2, 2)),
layers.Flatten(),
layers.Dropout(0.5),
layers.Dense(num_classes, activation="softmax"),
])
```
### Function-based Usage
```python
import keras
from keras import layers
from rt_activation import RTA
model = keras.Sequential([
keras.Input(shape=input_shape),
layers.Conv2D(32, kernel_size=(3, 3), activation=RTA),
layers.MaxPooling2D(pool_size=(2, 2)),
layers.Conv2D(64, kernel_size=(3, 3), activation=RTA),
layers.MaxPooling2D(pool_size=(2, 2)),
layers.Flatten(),
layers.Dropout(0.5),
layers.Dense(num_classes, activation="softmax"),
])
```
## Properties
- **Smooth**: Differentiable everywhere
- **Non-saturating**: Linear growth for large positive values
- **Bounded for negatives**: tanh behavior for negative inputs
- **Zero-centered**: Output can be negative
## License
MIT License
Raw data
{
"_id": null,
"home_page": null,
"name": "rt-activation",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.8",
"maintainer_email": null,
"keywords": "keras, tensorflow, activation, neural-networks, deep-learning",
"author": null,
"author_email": "Gaurav Pandey <gaurav.pandey1812@gmail.com>",
"download_url": "https://files.pythonhosted.org/packages/e1/0a/fe98e56dd287393e1582beb2c8026c3e8921420356422cb610d93d9e3ac8/rt_activation-1.0.0.tar.gz",
"platform": null,
"description": "# RT Activation\nA custom activation function for Keras/TensorFlow implementing Rectified Tangent Activation (RTA).\n\n## Formula\nf(x) = max(x, tanh(x))\n\n## Installation\n```bash\npip install rt-activation\n```\n## Usage\n\n### Simple Usage (String-based)\n```python\nimport keras\nfrom keras import layers\nimport rt_activation # This registers the activation function\n\nmodel = keras.Sequential([\n keras.Input(shape=input_shape),\n layers.Conv2D(32, kernel_size=(3, 3), activation=\"RTA\"),\n layers.MaxPooling2D(pool_size=(2, 2)),\n layers.Conv2D(64, kernel_size=(3, 3), activation=\"RTA\"),\n layers.MaxPooling2D(pool_size=(2, 2)),\n layers.Flatten(),\n layers.Dropout(0.5),\n layers.Dense(num_classes, activation=\"softmax\"),\n])\n```\n\n### Function-based Usage\n```python\nimport keras\nfrom keras import layers\nfrom rt_activation import RTA\n\nmodel = keras.Sequential([\n keras.Input(shape=input_shape),\n layers.Conv2D(32, kernel_size=(3, 3), activation=RTA),\n layers.MaxPooling2D(pool_size=(2, 2)),\n layers.Conv2D(64, kernel_size=(3, 3), activation=RTA),\n layers.MaxPooling2D(pool_size=(2, 2)),\n layers.Flatten(),\n layers.Dropout(0.5),\n layers.Dense(num_classes, activation=\"softmax\"),\n])\n```\n\n## Properties\n- **Smooth**: Differentiable everywhere\n- **Non-saturating**: Linear growth for large positive values\n- **Bounded for negatives**: tanh behavior for negative inputs\n- **Zero-centered**: Output can be negative\n\n## License\nMIT License\n",
"bugtrack_url": null,
"license": "MIT",
"summary": "Rectified Tangent Activation (RTA) function for Keras/TensorFlow",
"version": "1.0.0",
"project_urls": {
"Homepage": "https://github.com/gkp1812/rt-activation",
"Issues": "https://github.com/gkp1812/rt-activation/issues",
"Repository": "https://github.com/gkp1812/rt-activation"
},
"split_keywords": [
"keras",
" tensorflow",
" activation",
" neural-networks",
" deep-learning"
],
"urls": [
{
"comment_text": null,
"digests": {
"blake2b_256": "ac6439ff1a04f1153693f8fa1ca342f657d1177d19653aa706a106cdc93292f8",
"md5": "0c42f8af18c8bcd4ec6e362f8edd2070",
"sha256": "6fc0211a955a47433e516de5738291a0cbaa8cd930609d24f80a115403495625"
},
"downloads": -1,
"filename": "rt_activation-1.0.0-py3-none-any.whl",
"has_sig": false,
"md5_digest": "0c42f8af18c8bcd4ec6e362f8edd2070",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.8",
"size": 4240,
"upload_time": "2025-07-28T18:53:51",
"upload_time_iso_8601": "2025-07-28T18:53:51.317260Z",
"url": "https://files.pythonhosted.org/packages/ac/64/39ff1a04f1153693f8fa1ca342f657d1177d19653aa706a106cdc93292f8/rt_activation-1.0.0-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": null,
"digests": {
"blake2b_256": "e10afe98e56dd287393e1582beb2c8026c3e8921420356422cb610d93d9e3ac8",
"md5": "28607380c7da24b86890dede26930947",
"sha256": "9975fe053ec5ac271d466fbedaf3b0a126eb81604e26bce4c797221844002b12"
},
"downloads": -1,
"filename": "rt_activation-1.0.0.tar.gz",
"has_sig": false,
"md5_digest": "28607380c7da24b86890dede26930947",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.8",
"size": 3952,
"upload_time": "2025-07-28T18:53:52",
"upload_time_iso_8601": "2025-07-28T18:53:52.422094Z",
"url": "https://files.pythonhosted.org/packages/e1/0a/fe98e56dd287393e1582beb2c8026c3e8921420356422cb610d93d9e3ac8/rt_activation-1.0.0.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2025-07-28 18:53:52",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "gkp1812",
"github_project": "rt-activation",
"github_not_found": true,
"lcname": "rt-activation"
}
JSON
