Github: https://github.com/thursy
Academia: https://cycu.academia.edu/ThursySatriani
Scribd: https://www.scribd.com/user/30740323/Thursy-Satriani
If I’m not on social media, I am probably watching Mukbang, read books, debugging, or walking outside.
I don’t know if God is a cryptographer, but codes are all around us waiting to be deciphered. Some may take a thousand years for us to understand. Some may always be shrouded in mystery.
I wrote this down in my notebook but forget who said this
Published Paper
Day-ahead spatiotemporal wind speed forecasting using robust design-based deep learning neural network
The operations of power systems are becoming increasingly challenging due to the high penetration of renewable power generation, which is uncertain and stochastic. Highly accurate day-ahead renewable power forecasting helps operators in a dispatch center schedule power generations for conventional thermal generators. Efficient day-ahead renewable power forecasting requires information from widely separated locations. This paper proposes a novel day-ahead spatiotemporal wind speed forecasting method based on a convolutional neural network (CNN), which is an image-based deep learning method. This paper exploits a robust design based on Taguchi’s orthogonal array to determine the numbers of cascade/parallel layers, convolutions in each layer, kernels and hidden layers, as well as the kernel size, zero-padding and dropout ratio in the proposed CNN to ensure that the proposed CNN is insensitive to the seasonal characteristics of wind speeds while retaining high accuracy. In this paper, the day-ahead forecasting of wind speed at an offshore wind farm (Fuhai) near Taiwan is performed using historical wind speeds at seven sites in Taiwan, China, South Korea and the Philippines. Simulation results reveal that the proposed robust design-based CNN outperforms existing methods.
Medium-Writing
Docker Desktop CLI Essentials
This article provides a reference guide for Docker Desktop CLI commands. It covers essential concepts like Dockerfiles, images, and containers. It then dives into building images, running containers, and customizing them using environment variables. The guide is suitable for both beginners and experienced developers.
Managing Data in IBM Cloud
This article explains how to use IBM Cloud Object Storage to manage large amounts of data. Object storage organizes data into buckets like folders. The article provides a step-by-step guide to create a bucket and use Python code to upload and download data to and from the bucket. It also includes code for uploading entire directory structures.
Scheduling Streaming Insert ke BigQuery Menggunakan Serverless Option
The article discusses how to build an architecture using serverless services on Google Cloud Platform (GCP) to schedule script execution like Apache Airflow. It uses Cloud Scheduler to trigger a Pub/Sub topic, which triggers a Cloud Function that consumes an API and stores the data in a BigQuery table. The article highlights the benefits of serverless services like auto-scaling and cost-friendliness.
Dealing with Nominal, High Cardinality Data!
Categorical data like postal codes needs conversion to numerical form for machine learning. This article explores techniques like one-hot encoding (creating separate columns) and label encoding (assigning numbers) for both ranked (ordinal) and unranked (nominal) data. It emphasizes splitting data for training and testing before encoding, and using the training data’s mapping to encode the testing data.
How to Connect Redshift to Google Data Studio
Data Studio allows you to create visualizations from Redshift data. After acquiring Redshift credentials and adding Data Studio’s IP address to your Redshift security group, you can connect to the data in Data Studio and use a custom query to get the data you need for your visualizations.
3 Things you can do to deal with Imbalance Data Set
When working with data classification, imbalanced datasets can occur where one class has significantly more data than others. This can cause issues with training models. To address this, you can try techniques like downsampling the majority class, oversampling the minority class using SMOTE, or assigning class weights during model training.
Machine Learning di Google Cloud
Google Cloud Platform offers various tools for machine learning, including compute, storage, and pre-trained models or custom training with AutoML. This makes machine learning accessible to both beginners and experts.
Cloud Vision API — OCR with Notebook on GCP
Cloud Vision API is a Google Cloud service including the capability to do Optical Character Recognition (OCR). This tutorial will show how to use Vision API on a GCP Notebook.
Serverless Machine Learning GCP
This article discusses an overview on how to build a machine learning model in a serverless manner with GCP. The brief explanation about machine learning concepts and how to implement it using BigQuery Machine Learning or TensorFlow and Keras are also covered here.
Webinar & Workshop
Howdy! Here is list of my upcoming webinar and talks! You are welcome to join if you are interested. Mostly I will cover topic related to GenerativeAI, Machine Learning, IBM Cloud, and Google Cloud Platform.
Past events about Google Cloud….
Past events about Cloud Ace….