1d-convolution

Audio processing by using pytorch 1D convolution network

Deep learning approach for estimation of Remaining Useful Life (RUL) of an engine

A pytorch package for non-negative matrix factorization.

[深度应用]·首届中国心电智能大赛初赛开源Baseline（基于Keras val_acc: 0.88）

SeqNet: Code for the RA-L (ICRA) 2021 paper "SeqNet: Learning Descriptors for Sequence-Based Hierarchical Place Recognition"

This is a CNN based model which aims to automatically classify the ECG signals of a normal patient vs. a patient with AF and has been trained to achieve up to 93.33% validation accuracy.

In this work we propose two postprocessing approaches applying convolutional neural networks (CNNs) either in the time domain or the cepstral domain to enhance the coded speech without any modification of the codecs. The time domain approach follows an end-to-end fashion, while the cepstral domain approach uses analysis-synthesis with cepstral domain features. The proposed postprocessors in both domains are evaluated for various narrowband and wideband speech codecs in a wide range of conditions. The proposed postprocessor improves speech quality (PESQ) by up to 0.25 MOS-LQO points for G.711, 0.30 points for G.726, 0.82 points for G.722, and 0.26 points for adaptive multirate wideband codec (AMR-WB). In a subjective CCR listening test, the proposed postprocessor on G.711-coded speech exceeds the speech quality of an ITU-T-standardized postfilter by 0.36 CMOS points, and obtains a clear preference of 1.77 CMOS points compared to G.711, even en par with uncoded speech.

Models supported: ResNet, ResNetV2, SE-ResNet, ResNeXt, SE-ResNeXt [layers: 18, 34, 50, 101, 152] (1D and 2D versions with DEMO for Classification and Regression).

Models Supported: Inception [v1, v2, v3, v4], SE-Inception, Inception_ResNet [v1, v2], SE-Inception_ResNet (1D and 2D version with DEMO for Classification and Regression)

1D convolutional neural networks for activity recognition in python.

This repository provides the code used to create the results presented in "Global canopy height regression and uncertainty estimation from GEDI LIDAR waveforms with deep ensembles".

Predict blood pressure (systolic and diastolic) rythm from ECG and pulse oximeter signals

Classification models 1D Zoo - Keras and TF.Keras

用Tensorflow2.0進行外匯收盤價預測, 使用LSTM, 1d-CNN, Seq2Seq

Set of 1D CNN models to classify sound clips from the Urban Sound Classification dataset using Keras and Librosa

Play The Piano With Deep Learning 用深度学习弹钢琴 2019-5-22

1-D convolution implementation using Python and CUDA

Models Supported: VGG11, VGG13, VGG16, VGG16_v2, VGG19 (1D and 2D versions with DEMO for Classification and Regression).

IDH and TERTp mutation classification in gliomas using 1D-CNN with MRS data.

Impulse Classification Network (ICN) for video Head Impulse Test

Anti-hydrogen detection using CNNs from ASACUSA experiment

Protein structure prediction using 1d CNN and GRU 🧬

Set of 2D & 1D CNN models to classify images of handwritten numbers from the MNIST dataset using Keras.

1-dimensional convolution implementation using C++ and CUDA

Bengali Newses are classified in six catagories. This is done by first extracting the semantics of Bengali words using word2vec. Then using those semantics, all the news are classified. Bengali NLP resources are not very rich compared to other languages. This is a complete project that includes Bengali word embedding, data cleaning using word stemming, stop words, punctuation removal, tokenization and finally training the model using Neural Networks having convolutional layers.

Movie ratings prediction

This is the repo that will be used to store the code used for the Intel / IBACs AI technical workshop hosted at the University of Connecticut.

Model for translating speech to text. This is similar to using amazon alexa amongst many other products

a CNN that used word embeddings from GloVe to classify toxic comments

1DConvNet applied to room occupancy detection based on data from several environment sensors. Data courtesy of the UCI Machine Learning Repository.