pydeepflow 0.1.3

A deep learning package optimized for performing Deep Learning Tasks, easy to learn and integrate into projects

---

PyDeepFlow Documentation

Overview

PyDeepFlow is a deep learning package optimized for performing deep learning tasks. It is designed to be easy to learn and integrate into various projects. The package provides a set of utilities for model building, training, and evaluation, making it a suitable choice for both beginners and experienced practitioners.

Table of Contents

• Features
• Installation
• Usage
  • Basic Example
  • Model Training
  • Evaluating the Model
• API Reference
  • Modules
  • Classes and Functions
• Examples
• Contributing
• License
• Contact

Features

• Easy-to-use API for deep learning tasks.
• Support for various activation functions, loss functions, and optimizers.
• Compatibility with Jupyter notebooks for interactive usage.
• Flexible model training and evaluation utilities.
• Extensive documentation and examples for quick start.

Installation

To install PyDeepFlow, you can use pip. Ensure you have Python 3.6 or higher installed. You can install the package from PyPI using the following command:

pip install pydeepflow

To install the package along with its dependencies, you can run:

pip install -r requirements.txt

Usage

Basic Example

Here’s a quick example of how to use PyDeepFlow to create and train a simple model.

from pydeepflow.model import Model
from pydeepflow.losses import MeanSquaredError
from pydeepflow.activations import ReLU
import numpy as np

# Generate some random data
X = np.random.rand(100, 10)
y = np.random.rand(100, 1)

# Create a model
model = Model()

# Add layers to the model
model.add_layer(units=64, activation=ReLU())
model.add_layer(units=1, activation=None)

# Compile the model
model.compile(loss=MeanSquaredError(), optimizer='adam')

# Train the model
model.fit(X, y, epochs=10, batch_size=16)

Model Training

To train a model, you can call the fit method on your model instance. Here’s a more detailed explanation of the parameters:

• X: Input data (numpy array).
• y: Target data (numpy array).
• epochs: Number of times to iterate over the training data.
• batch_size: Number of samples per gradient update.

model.fit(X, y, epochs=20, batch_size=32)

Evaluating the Model

After training the model, you can evaluate its performance using the evaluate method.

loss = model.evaluate(X, y)
print(f'Model Loss: {loss}')

API Reference

Modules

• activations: Contains various activation functions.
• losses: Contains loss functions to measure the model performance.
• model: Defines the model structure and training procedures.
• device: Provides functionalities related to device management (e.g., GPU support).

Classes and Functions

Model Class

class Model:
    def __init__(self):
        # Initializes the model
        pass

    def add_layer(self, units, activation):
        # Adds a layer to the model
        pass

    def compile(self, loss, optimizer):
        # Compiles the model
        pass

    def fit(self, X, y, epochs, batch_size):
        # Trains the model on the provided data
        pass

    def evaluate(self, X, y):
        # Evaluates the model on the provided data
        pass

Loss Functions

class MeanSquaredError:
    def __call__(self, y_true, y_pred):
        # Calculates mean squared error
        pass

Activation Functions

class ReLU:
    def __call__(self, x):
        # Applies ReLU activation function
        pass

Examples

Explore the examples/ directory for various Jupyter notebooks demonstrating how to use PyDeepFlow for different deep learning tasks:

• Iris_Load_and_Save.py: Load and save Iris dataset.
• Iris_Multi_Class.py: Perform multi-class classification on the Iris dataset.
• Titanic.ipynb: Analyze Titanic survival data and build a predictive model.

References

1. Neural Networks and Deep Learning:

   • Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep Learning. MIT Press. Link

2. Python for Data Analysis:

   • McKinney, W. (2017). Python for Data Analysis: Data Wrangling with Pandas, NumPy, and IPython. O'Reilly Media.

3. Scikit-learn Documentation:

   • Scikit-learn. (n.d.). Scikit-learn: Machine Learning in Python. Link

4. TQDM Documentation:

   • TQDM. (n.d.). TQDM: A fast, extensible progress bar for Python and CLI. Link

5. Colorama Documentation:

   • Colorama. (n.d.). Colorama: Simple cross-platform print formatting. Link

Contributing

We welcome contributions to PyDeepFlow. If you'd like to contribute, please follow these steps:

1. Fork the repository on GitHub.
2. Create a new branch for your feature or bug fix.
3. Make your changes and commit them.
4. Push your branch and create a pull request.

Guidelines

• Ensure that your code adheres to the existing code style.
• Include tests for any new functionality.
• Update documentation if necessary.

License

PyDeepFlow is licensed under the MIT License. See the LICENSE file for more details.

Contact

For questions or inquiries, please contact:

• Author: Ravin D
• Email: [e-mail]
• GitHub: GitHub Profile

---