Deep Learning

Multiclass bearing fault classification using features learned by a deep neural network

Accurate classification of fault is important for condition based maintenance (CBM) applications. Two main approaches commonly used for fault classification are model-based and …

avatar
Biswajit Sahoo

IndexedSlices in Tensorflow

What are `IndexedSlices` and why are they get used in Tensorflow?

avatar
Biswajit Sahoo

Tensorflow 2 code for Attention Mechanisms chapter of Dive into Deep Learning (D2L) book

Implementing attention mechanisms, multi-head attention, transformer architecture, etc. from scratch in Tensorflow.

avatar
Biswajit Sahoo

Reading multiple files in Tensorflow 2 using Sequence

Create a data pipeline by incrementally reading thousands of `csv` files using Tensorflow `Sequence`.

avatar
Biswajit Sahoo

Reading multiple csv files in PyTorch

Create a data pipeline by incrementally reading thousands of `csv` files in PyTorch.

avatar
Biswajit Sahoo

Data-Driven Remaining Useful Life (RUL) Prediction

Aim of this project is to produce reproducible results in condition monitoring. We will apply some of the standard machine learning techniques to publicly available datasets and …

Efficiently reading multiple files in Tensorflow 2

Use of `map` and `prefetch` combined with generators can speed up data ingestion.

avatar
Biswajit Sahoo

Reading multiple files in Tensorflow 2

Create a data pipeline by incrementally reading thousands of `csv` files in Tensorflow.

avatar
Biswajit Sahoo

Using Python Generators

Understanding python generators with simple examples.

avatar
Biswajit Sahoo

Data-Driven Machinery Fault Diagnosis

Aim of this project is to produce reproducible results in condition monitoring. We will apply some of the standard machine learning techniques to publicly available machinery …