Metadata-Version: 2.4
Name: comfi_fast_grnn_tensorflow
Version: 0.0.1
Summary: A PyTorch implementation of Fast ULCNet
Author-email: Nicolas Arrieta Larraza <NIAL@bang-olufsen.dk>, Niels de Koeijer <NEMK@bang-olufsen.dk>
License: MIT
Project-URL: Homepage, https://github.com/narrietal/Fast-ULCNet
Requires-Python: >=3.10
Description-Content-Type: text/markdown
Requires-Dist: tensorflow>=2.14.0
Requires-Dist: tensorflow-estimator>=2.14.0
Requires-Dist: libsegmenter==1.0.4
Requires-Dist: CRM_tensorflow==0.1.6

# fast-ulcnet-tensorflow
Implements Comfi-FastGRNN in tensorflow.

## Usage

You can use the `ComfiFastGRNN` layer just like any standard Keras RNN layer (e.g., `LSTM`, `GRU`). It supports the Sequential and Functional APIs.

### Basic Implementation
The simplest way to use the layer with default settings:

```python
import tensorflow as tf
from comfi_fast_grnn_tensorflow import ComfiFastGRNN  

# Define a Sequential model
model = tf.keras.Sequential([
    # Input shape: (Timesteps, Features)
    tf.keras.layers.Input(shape=(100, 32)), 
    
    # Comfi-FastGRNN layer
    ComfiFastGRNN(
        units=64, 
        return_sequences=False
    ),
    
    tf.keras.layers.Dense(10, activation='softmax')
])

model.summary()
```
