issues
search
behzadhaki
/
GrooveTransformer
Variational version of Monotonic Groove Transformer
14
stars
5
forks
source link
readme
List of Contents
Guides for using the repository can be found here
[Chapter 0]()
A. ENVIRONMENT SETUP
Required Packages
CPU Installation
venv Installation
Anaconda Installation
GPU Installation
Local GPU Installation
HPC Cluster GPU Installation
Anaconda Installation
B. HPC Cluster Guide
Getting Started
Accounts
Accessing the Clusters
Using the Clusters
Resources
Interactive Sessions
Submitting Jobs
Available Software
Monitoring or Cancelling Jobs
[Chapter 1]()
A. Data Representation & Datasets
Introduction
Data Representation
HVO_Sequence
Example Code
Datasets
Groove Midi Dataset
Load dataset as a dictionary
Extract
HVO_Sequence
objects from dataset dictionaries
_Load GMD Dataset in
HVO_Sequence
format using a single command !!!_
B. HVO Sequence
Basic Attributes
Beat Division Factors
Drum Mapping
Grid Attributes
Metadata
HVO: Piano-roll Score
Simple Usage
Built-in Tools
Multi-Segment Scores
Chapter 2 - Models
A. Groove Transformer
Introduction
Instantiating a Model
BasicGrooveTransformer.GrooveTransformer
BasicGrooveTransformer.GrooveTransformerEncoder
Storing a Model
Loading a Stored Model
Generation using a Model
B. Variational Groove transformer
Introduction
Model Description
Network Architecture
loss functions
Training Parameters
MonotonicGrooveVAE.GrooveTransformerEncoderVAE
Instantiation
Storing
Loading
Pretrained Versions
Generation
Chapter 3 - Evaluation Tools
A.
GrooveEvaluator
Part A1
Prepapre the data used for Evaluation
Initialization
Preparing Predictions
Get Ground Truth Samples
Pass Samples to Model
Add Predictions to Evaluator
Saving and Loading
Part A2
Accessing Evaluation Results
Results as Dictionaries or Pandas.DataFrame
Rendering Results as Bokeh Plots
Rendering Piano Rolls/Audio/Midi
Compiling Plots for Logging
B.
MultiSetEvaluator
Prepapre the sets used for cross comparison
Initialization
Saving and Loading
Available Analyzers
Inter-Intra Analysis (raw statistics, distribution plots and KL/OA Plots)
Hit, Velocity, Offset Analysis
Compiling Results
Chapter 4 - WANDB
Chapter 5 - Training