Tag: Neural Network

  • Sales Forecasting: tslearn, Random Walk, Holt-Winters, SARIMAX, GARCH, Prophet, and LSTM

    Sales Forecasting: tslearn, Random Walk, Holt-Winters, SARIMAX, GARCH, Prophet, and LSTM

    The data science project involves evaluating various sales forecasting algorithms in Python using a Kaggle time-series dataset. The forecasting algorithms include tslearn, Random Walk, Holt-Winters, SARIMA, GARCH, Prophet, LSTM and Di Pietro’s Model. The goal is to predict next month’s sales for a list of shops and products, which slightly changes every month. The best…

  • Hugging Face NLP, Streamlit, PyGWalker, TF & Gradio App

    Hugging Face NLP, Streamlit, PyGWalker, TF & Gradio App

    Table of Contents Streamlit/Dash/Jupyter PyGWalker EDA Demo PyGWalker and Dash — Creating a Data Visualization Dashboard In Less Than 20 Lines of Code PyGWalker Test PyGWalker Tutorial: A Tableau-Like Python Library for Interactive Data Exploration and Visualization PyGWalker: A Python Library for Visualizing Pandas Dataframes You’ll Never Walk Alone: Use Pygwalker to Visualize Data in…

  • Returns-Volatility Domain K-Means Clustering and LSTM Anomaly Detection of S&P 500 Stocks

    Returns-Volatility Domain K-Means Clustering and LSTM Anomaly Detection of S&P 500 Stocks

    This study aims to implement and evaluate the K-means algorithm for ranking/clustering S&P 500 stocks based on average annualized return and volatility. The second goal is to detect anomalies in the best performing S&P 500 stocks using the Isolation Forest algorithm. Additionally, anomalies in the S&P 500 historical stock price time series data will be…

  • Real-Time Anomaly Detection of NAB Ambient Temperature Readings using the TensorFlow/Keras Autoencoder

    Real-Time Anomaly Detection of NAB Ambient Temperature Readings using the TensorFlow/Keras Autoencoder

    The content covers a detailed guide on implementing anomaly detection in time series data using autoencoders. The tutorial utilizes Python and real-world temperature dataset from Numenta Anomaly Benchmark (NAB). Following the Python workflow, the algorithm imports required libraries, performs anomaly detection, and visualizes anomalies. A trained autoencoder model identifies anomalies, with Precision, Recall, and F1…

  • NVIDIA Returns-Drawdowns MVA & RNN Mean Reversal Trading

    NVIDIA Returns-Drawdowns MVA & RNN Mean Reversal Trading

    The study presents a machine learning-focused analytical approach to optimize NVIDIA’s stock performance using moving average crossovers and aims at comparing the outcomes with simple RNN mean reversal trading strategies. The steps taken involve preparing the stock data, calculating moving averages and drawdowns, plotting heatmaps of returns and drawdowns, and predicting returns and cumulative returns…

  • Image Based Fast Forest Fire Detection with TensorFlow

    Image Based Fast Forest Fire Detection with TensorFlow

    A recent study showcases the use of artificial intelligence (AI) and deep learning (DL) for efficient wildfire prediction and management. Utilizing a fast DL approach based on the TensorFlow Convolution Neural Network (CNN) algorithm, researchers trained models to distinguish between fire and non-fire images using a public-domain dataset. The implemented system predicted fires accurately and…

  • Improved Multiple-Model ML/DL Credit Card Fraud Detection: F1=88% & ROC=91%

    Improved Multiple-Model ML/DL Credit Card Fraud Detection: F1=88% & ROC=91%

    In 2023, the global card industry is projected to suffer $36.13 billion in fraud losses. This has necessitated a priority focus on enhancing credit card fraud detection by banks and financial organizations. AI-based techniques are making fraud detection easier and more accurate, with models able to recognize unusual transactions and fraud. The post discusses a…

  • Early Heart Attack Prediction using ECG Autoencoder and 19 ML/AI Models with Test Performance QC Comparisons

    Early Heart Attack Prediction using ECG Autoencoder and 19 ML/AI Models with Test Performance QC Comparisons

    Table of Contents Embed Socials: ECG Autoencoder Let’s set the working directory YOURPATH import osos.chdir(‘YOURPATH’)os. getcwd() and import the following libraries import tensorflow as tfimport matplotlib.pyplot as pltimport numpy as npimport pandas as pd from tensorflow.keras import layers, lossesfrom sklearn.model_selection import train_test_splitfrom tensorflow.keras.models import Model Let’s read the input dataset df = pd.read_csv(‘ecg.csv’, header=None) Let’s…

  • Using AI/ANN AUC>90% for Early Diagnosis of Cardiovascular Disease (CVD)

    Using AI/ANN AUC>90% for Early Diagnosis of Cardiovascular Disease (CVD)

    The project utilizes AI-driven cardiovascular medicine with a focus on early diagnosis of heart disease using Artificial Neural Networks (ANN). Aiming to improve early detection of heart issues, the project processed a dataset of 303 patients using Python libraries and conducted extensive exploratory data analysis. A Sequential ANN model was subsequently built, revealing excellent performance…

  • Overview of AWS Tech Portfolio 2023

    Overview of AWS Tech Portfolio 2023

    This summary focuses on the extensive capabilities of Amazon Web Services (AWS) by 2023, highlighting its 27% year-on-year growth and a net sales increase to $127.1 billion. AWS emerges as the top cloud service provider, offering over 200 services including compute, storage, databases, networking, AI, and machine learning. It is constantly expanding operations, having opened…

  • AI-Driven Object Detection & Segmentation with Meta Detectron2 Deep Learning

    AI-Driven Object Detection & Segmentation with Meta Detectron2 Deep Learning

    The post introduces Detectron2, a powerful object detection and image recognition platform developed by Facebook AI Research (FAIR). It discusses the platform’s applications in computer vision research and production, as well as its capabilities such as panoptic segmentation and Densepose. The post also covers aspects of using Detectron2, including installation, model training, and inference, and…

  • Comparative ML/AI Performance Analysis of 13 Handwritten Digit Recognition (HDR) Scikit-Learn Algorithms with PCA+HPO

    Comparative ML/AI Performance Analysis of 13 Handwritten Digit Recognition (HDR) Scikit-Learn Algorithms with PCA+HPO

    Featured Photo by Torsten Dettlaff on Pexels The article consists of the following three parts: 3. Unsupervised ML using the Principal Component Analysis (PCA) for the dimensionality reduction within Parts 1 and 2. Our main goal is to build a text and graphics report comparing the main scikit-learn classification metrics: accuracy_score, classification_report (precision, recall, and…

  • Multi-Label Keras CNN Image Classification of MNIST Fashion Clothing

    Multi-Label Keras CNN Image Classification of MNIST Fashion Clothing

    Machine learning and deep learning are invaluable in optimizing supply chain operations in fashion retail. Even smaller retailers are leveraging ML algorithms to meet customer demands. Neural network models, particularly Convolution Neural Networks (CNN) are used to classify clothing images, like the Fashion-MNIST dataset, with high accuracy. Hyperparameter optimization using GridSearchCV and Nadam optimizer are…

  • AI-Based ECG Recognition – EOY ’22 Status

    AI-Based ECG Recognition – EOY ’22 Status

    Featured Photo by cottonbro studio on pexels. Electrocardiography (ECG) is the method most often used to diagnose cardiovascular diseases. The recent study demonstrates that an AI is capable of automatically diagnosing the abnormalities indicated by an ECG. In this post we will review and illustrate how AI applies to ECG analysis to outperform traditional ECG analysis.…

  • 99% Accurate Breast Cancer Classification using Neural Networks in TensorFlow

    99% Accurate Breast Cancer Classification using Neural Networks in TensorFlow

    Breast cancer is a significant global health concern, affecting 12% of women. Machine Learning and Artificial Intelligence techniques play a crucial role in early diagnosis using image features. The study demonstrates a successful Neural Network model for breast cancer classification, achieving 98% accuracy and 98% F1-score. Multiple metrics confirm the model’s efficiency.

  • DL-Assisted ECG/EKG Anomaly Detection using LSTM Autoencoder

    DL-Assisted ECG/EKG Anomaly Detection using LSTM Autoencoder

    This project implements an ECG anomaly detection framework using an LSTM Autoencoder to accurately identify abnormal ECG events. It trains the autoencoder on normal rhythms, using reconstruction errors to identify anomalies. The proposed method aims to improve abnormal ECG detection, as demonstrated by test results on the ECG5000 dataset, providing valuable information for patient health…