Skip to content

Audio data preprocessing to classify urban sound using deep learning techniques.

Notifications You must be signed in to change notification settings

Rekidiang2/ml06_urban_sound_classification

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

6 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Urban Sound Classification

photo cover

competition

Exercises

Dataset

freesound.org

Paper with code

In this project a dataset with music of different styles such us jazz, classic ...

Contents

  1. Project Structure
  2. Prosess
  3. How to run
  4. Deployment (Real world Use)
  5. To improve
  6. About Me

1. Project Structure

Data

  • ├── diabetes.csv
  • ├── cleaned_data.csv
  • ├── scaled_data.csv
  • ├── data_documentation.pdf

analysis_and_training

  • ├── 1_data_preprocessing
  • ├── 2_urban_sound_Classification_with_ANN.ipynb
  • ├── 3_urban_sound_Classification_with_CNN.ipynb
  • ├── 4_urban_sound_Classification_with_RNN.ipynb
  • ├── helpers.py
  • ├── models_config.py

figures

  • ├── contains graph and figures

models

  • ├── contains trained models

images

  • ├── contains images used in this images
environment
requirements.txt
gitignore
report (pdf & ppt)

2. Process

  • step1 : Importing Packages
  • step2 : Loading the data
  • step3 : Exploratory Data Analysis (EDA)
  • step4 : Data Preparation
  • step5 : Build and Train the model
  • step6 : Model prediction and Evaluation
  • step7 : Model Improvement
  • Hyperparameter Tuning, Features Selection and Features Ingeneering
  • step8 : Model Deployment

3. How to run

N.B : python 3.7 is recommended

3.1. CLONE PROJECT DIRECTORY

  • $ git clonehttps://github.com/Rekidiang2/au03_urban_sound_classification.git
  • $ cd au03_urban_sound_classification

3.2. CREATE & ACTIVATE VIRTUAL ENVIRONMENT

3.2.1. WITH PIP and VENV

(Windows)
  • $ python -m venv au03_venv
  • $ au03_venv\Scripts\activate (<= Activate virtual Environment)
  • $ deactivate (<= Deactivate virtual Environment)
  • $ pip install -r requirements.txt
  • Set VIRTUAL ENVIRONMENT as KERNEL :
    • $ python -m ipykernel install --user --name au03_venv --display-name "au03_kernel"
  • $ jupyter notebook
  • Go to Kernel -> Change kernel -> au03_kernel
(MasOS || LINUX)
  • $ python3 -m venv au03_venv
  • $ source au03_venv/bin/activate (<= Activate virtual Environment)
  • $ deactivate (<= Deactivate virtual Environment)
  • $ pip install -r requirements.txt
  • Set VIRTUAL ENVIRONMENT as KERNEL :
    • $ python -m ipykernel install --user --name au03_venv --display-name "au03_kernel"
  • $ jupyter notebook
  • Go to Kernel -> Change kernel -> au03_kernel

3.2.2. WITH CONDA

  • Verify if you have conda installed ($conda --version) if not go to anconda or miniconda to download and install it

  • $ conda create -n au03_venv python=3.7

  • $ conda activate au03_venv (<= Activate virtual Environment)

  • $ conda deactivate (<= Deactivate virtual Environment)

  • Set VIRTUAL ENVIRONMENT as KERNEL :

    • $ python -m ipykernel install --user --name au03_venv --display-name "au03_kernel"
  • $ jupyter notebook

  • Go to Kernel -> Change kernel -> au03_kernel

Manage kernel

  • $ jupyter kernelspec list (<= list all ipykernel in your system)
  • $ jupyter kernelspec uninstall au03_kernel (<= Delete the ipykernel in your system)

4. Deployment (Real world Use)

5. To improve

  • put link for real world app after deployment in heroku

6. About Me


I'm Data Analyst, Data Scientist and Web Developer. Data and technology passionate person, Artificial Intelligence enthusiast.

My Website Click Here

Social Network

alt text alt text alt text alt text

About

Audio data preprocessing to classify urban sound using deep learning techniques.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published