keunwoochoi / Transfer_learning_music
Transfer learning for music classification and regression tasks
Stars: ✭ 228
Labels
Projects that are alternatives of or similar to Transfer learning music
Pydata Book
Materials and IPython notebooks for "Python for Data Analysis" by Wes McKinney, published by O'Reilly Media
Stars: ✭ 16,386 (+7086.84%)
Mutual labels: jupyter-notebook
Question Generation
Generating multiple choice questions from text using Machine Learning.
Stars: ✭ 227 (-0.44%)
Mutual labels: jupyter-notebook
Coronavirus Epidemic Covid 19
👩🏻⚕️Covid-19 estimation and forecast using statistical model; 新型冠状病毒肺炎统计模型预测 (Jan 2020)
Stars: ✭ 228 (+0%)
Mutual labels: jupyter-notebook
Alphatools
Quantitative finance research tools in Python
Stars: ✭ 226 (-0.88%)
Mutual labels: jupyter-notebook
Statistics python codes
《统计学原理实验教程(Python)》书中代码实现。尽可能加注释,力求代码的可复用性。
Stars: ✭ 230 (+0.88%)
Mutual labels: jupyter-notebook
Deeplearning Models
A collection of various deep learning architectures, models, and tips
Stars: ✭ 14,654 (+6327.19%)
Mutual labels: jupyter-notebook
Snap N Eat
Food detection and recommendation with deep learning
Stars: ✭ 229 (+0.44%)
Mutual labels: jupyter-notebook
Functional intro to python
[tutorial]A functional, Data Science focused introduction to Python
Stars: ✭ 228 (+0%)
Mutual labels: jupyter-notebook
Pytorch Transformers Classification
Based on the Pytorch-Transformers library by HuggingFace. To be used as a starting point for employing Transformer models in text classification tasks. Contains code to easily train BERT, XLNet, RoBERTa, and XLM models for text classification.
Stars: ✭ 229 (+0.44%)
Mutual labels: jupyter-notebook
Gpt2bot
Your new Telegram buddy powered by transformers
Stars: ✭ 228 (+0%)
Mutual labels: jupyter-notebook
Satellite analysis
Analysis scripts of things related to satellites
Stars: ✭ 228 (+0%)
Mutual labels: jupyter-notebook
Deep Learning In Production
Develop production ready deep learning code, deploy it and scale it
Stars: ✭ 216 (-5.26%)
Mutual labels: jupyter-notebook
Cd4ml Workshop
Repository with sample code and instructions for "Continuous Intelligence" and "Continuous Delivery for Machine Learning: CD4ML" workshops
Stars: ✭ 229 (+0.44%)
Mutual labels: jupyter-notebook
Data
Data and code behind the articles and graphics at FiveThirtyEight
Stars: ✭ 15,241 (+6584.65%)
Mutual labels: jupyter-notebook
Scnn
Segment-CNN: A Framework for Temporal Action Localization in Untrimmed Videos via Multi-stage CNNs
Stars: ✭ 228 (+0%)
Mutual labels: jupyter-notebook
Predicting winning teams
This is the code for "Predicting the Winning Team with Machine Learning" by Siraj Raval on Youtube
Stars: ✭ 229 (+0.44%)
Mutual labels: jupyter-notebook
Dsnd term1
Contains files related to content and project of DSND
Stars: ✭ 229 (+0.44%)
Mutual labels: jupyter-notebook
Weatherbench
A benchmark dataset for data-driven weather forecasting
Stars: ✭ 227 (-0.44%)
Mutual labels: jupyter-notebook
transfer_learning_music
Repo for paper "Transfer learning for music classification and regression tasks" by Keunwoo Choi et al.
Mode 1/2. To use the pre-trained convnet feature extractor
For your own music/audio-related work.
Prerequisites (Same as mode 2 except datasets)
- Theano. I used version 0.9.0 but should work with some similar versions.
-
Keras 1.2.2 (OLD ONE!) (NOT THE MOST RECENT VERSION)
- set
image_dim_ordering : th
in~/keras/keras.json
- set
backend : theano
, too.
- set
- Kapre OLD VERSION for OLD KERAS In short,
$ pip install theano==0.9
$ pip install keras==1.2.2
$ git clone https://github.com/keunwoochoi/kapre.git
$ cd kapre
$ git checkout a3bde3e
$ python setup.py install
Usage
$ python easy_feature_extraction.py audio_paths.txt some/path/features.npy
where audio_path.txt
is line-by-line audio paths and some/path/features.npy
is the path to save the result.
E.g., audio_path.txt
:
blah/a.mp3
blahblah/234.wav
some/other.c.mp3
Then load the .npy
file. The features are size of (num_songs, 160)
.
Mode 2/2. To reproduce the paper
Prerequisites
-
Download datasets:
-
Prerequisites
- Theano. I used version 0.9.0 but should work with some similar versions.
-
Keras 1.2.2 (OLD ONE!) (NOT THE MOST RECENT VERSION)
- set
image_dim_ordering : th
in~/keras/keras.json
- set
backend : theano
, too.
- set
- Kapre OLD VERSION for OLD KERAS by
$ git clone https://github.com/keunwoochoi/kapre.git
$ cd kapre
$ git checkout a3bde3e
$ python setup.py install
- Optionally,
Sckikt learn, Pandas, Numpy
,.. for your convenience.
Usage
-
0. main_prepare_many_datasets.ipynb
: prepare dataset, pre-processing -
1. feature extraction for 6 tasks.ipynb
: feature extraction (MFCC and convnet features) -
2_main_knn_svm_transfer
: Do SVM -
3. knn and svm (with AveragePooling) results plots
: Plot results
Appendix
Links
- Train/valid/test split of MSD that I used for the training
- Paper: arXiv 1703.09179, Transfer Learning for Music Classification and Regression tasks
- Blog article
Citation:
@inproceedings{choi2017transfer,
title={Transfer learning for music classification and regression tasks},
author={Choi, Keunwoo and Fazekas, George and Sandler, Mark and Cho, Kyunghyun},
booktitle={The 18th International Society of Music Information Retrieval (ISMIR) Conference 2017, Suzhou, China},
year={2017},
organization={International Society of Music Information Retrieval}
}
Note that the project description data, including the texts, logos, images, and/or trademarks,
for each open source project belongs to its rightful owner.
If you wish to add or remove any projects, please contact us at [email protected].