OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Models

Deep Complex Networks

Deep Complex Networks is a Python repository that reproduces experiments from the Deep Complex Networks paper, with scripts for computer-vision training and MusicNet music-model experiments.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Deep Complex Networks — OpenViST project profile in AI Models.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use the repository when you want code and documented command-line workflows for reproducing the paper's computer-vision experiments or training its MusicNet models.

Who it is for

  • Researchers and practitioners reproducing the paper's computer-vision experiments.
  • Users training shallow or deep convolutional models on MusicNet.
  • Developers exploring real- and complex-valued neural-network models.

Documented uses

  • Train real- or complex-valued models for computer-vision experiments.
  • Prepare MusicNet audio data by resampling it from 44.1 kHz to 11 kHz.
  • Train shallow or deep real- and complex-valued MusicNet convolutional models.
  • Visualize MusicNet predictions and precision-recall results in the supplied notebook.
Before you choose

Requirements & limits

The public GitHub repository specifies an MIT license.

  • Computer-vision and music experiments have different listed dependency sets.
  • MusicNet training requires obtaining the dataset and resampling it before running the listed training commands.
  • Computer-vision training accepts real or complex model selection and exposes additional training settings through its command help.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Prepare MusicNet and train a model

Resample the MusicNet dataset and run a shallow or deep MusicNet training script.

Before you start

  • Python environment with scipy, sklearn, intervaltree, resampy and mimir installed.
  • MusicNet dataset downloaded to data/musicnet.npz.

Steps

  1. Resample the dataset with resample.py data/musicnet.npz data/musicnet_11khz.npz 44100 11000.

  2. For a shallow real-valued model, run train.py shallow_model --in-memory --model=shallow_convnet --local-data data/musicnet_11khz.npz; for a shallow complex-valued model, use train.py shallow_complex_model --in-memory --model=complex_shallow_convnet --complex --local-data data/musicnet_11khz.npz.

  3. For a deep real-valued model, run train.py deep_model --in-memory --model=deep_convnet --fourier --local-data data/musicnet_11khz.npz; for a deep complex-valued model, use train.py deep_complex_model --in-memory --model=complex_deep_convnet --fourier --complex --local-data data/musicnet_11khz.npz.

  4. Open notebooks/visualize_musicnet.ipynb to view the supplied prediction and precision-recall visualizations.

Expected outcome

A training run using a selected MusicNet convolutional model and a notebook for visualizing MusicNet predictions and precision-recall results.

Guide limitations

  • The README assumes the dataset is available locally before resampling and training.
  • It does not specify training duration, compute requirements or expected performance.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
7 November 2026
Original website title
ChihebTrabelsi/deep_complex_networks · Official homepage title · retained in its source language
How we review sourcesSuggest a correction