​Most Effective Way To Implement Radial Basis Function Neural Network for Classification Problem

🔗 Most Effective Way To Implement Radial Basis Function Neural Network for Classification Problem
How to use K-Means Clustering along with Linear regression to classify images

​Big Data Analysis for Bioinformatics and Biomedical Discoveries

Наш телеграм канал - tglink.me/ai_machinelearning_big_data

🔗 Открыть в Telegram



📝 Big Data Analysis for Biomedical Discoveries (en).pdf - 💾6 399 456

​Resolving the Scope of Speculation and Negation using Transformer-Based Architectures

🔗 Resolving the Scope of Speculation and Negation using Transformer-Based Architectures
Speculation is a naturally occurring phenomena in textual data, forming an integral component of many systems, especially in the biomedical information retrieval domain. Previous work addressing cue detection and scope resolution (the two subtasks of speculation detection) have ranged from rule-based systems to deep learning-based approaches. In this paper, we apply three popular

​Fast Neural Network Adaptation via Parameter Remapping and Architecture Search

https://github.com/JaminFong/FNA

Paper: https://arxiv.org/abs/2001.02525v1

🔗 JaminFong/FNA
Fast Neural Network Adaptation via Parameter Remapping and Architecture Search - JaminFong/FNA

​👨‍🦱 DeeperForensics-1.0: A Large-Scale Dataset for Real-World Face Forgery Detection

Code: https://github.com/EndlessSora/DeeperForensics-1.0

Paper: https://arxiv.org/abs/2001.03024v1

Наш телеграм канал - tglink.me/ai_machinelearning_big_data

🔗 EndlessSora/DeeperForensics-1.0
Code, model and data of DeeperForensics-1.0 will be made publicly available here. - EndlessSora/DeeperForensics-1.0

Знакомство с машинным обучением на бесплатном интенсиве от Skillbox — отличный шанс начать карьеру в Data Science и стать востребованным специалистом.

Регистрируйся по ссылке: ▶https://clc.to/rxEv9g

Всего три дня занятий — с 13 по 15 января, и ты откроешь себе дверь в профессию будущего!

💡 Интенсив проведёт Михаил Овчинников, главный методист технического направления Skillbox. Вместе с ним ты создашь искусственный интеллект, освоишь Python и Machine Learning с нуля.

​Real or Not? NLP with Disaster Tweets — EDA

🔗 Real or Not? NLP with Disaster Tweets — EDA
Getting started with NLP

​HybridPose: 6D Object Pose Estimation under Hybrid Representations

https://github.com/chensong1995/HybridPose

https://arxiv.org/abs/2001.01869v1

Наш телеграм канал - tglink.me/ai_machinelearning_big_data

🔗 chensong1995/HybridPose
Implementation of HybridPose: 6D Object Pose Estimation under Hybrid Representation - chensong1995/HybridPose

​What is Stationarity in Time Series and why should you care

🔗 What is Stationarity in Time Series and why should you care
Time Series are Everywhere — Make sure you know how to Handle them

​Accelerate Model Training With Batch Normalization

🔗 Accelerate Model Training With Batch Normalization
What is batch normalization and how does it work? What enables it to achieve faster training? This post aims to answer the above questions

🎥 React with Python Flask and Machine Learning/Vision - THIS IS A TEST!
👁 1 раз ⏳ 5621 сек.
In this video I check out a blog I found where Prediction is used inside a python flask and react app on github, and I also setup a starter github with Python Flask using Visual Studio and React JS as a starting point to do some (non-standard) architectural integration using a module blueprint approach. For this, there will be a second and third video to work through the integration points. Fun hacking on it, it was a litle long, but its cool, the next one will be even better :)

🎥 Machine Learning for Beginners - Supervised vs. Unsupervised Learning
👁 1 раз ⏳ 711 сек.
Welcome back to this series on Machine Learning for Beginners! This video will cover the different types of Machine Learning algorithms - Supervised Learning, Unsupervised Learning, Semisupervised Learning and finally Reinforcement Learning. We're almost to the coding portion - stay tuned!

► JOIN THECODEX TODAY! https://www.thecodex.me/
► FACEBOOK: https://www.facebook.com/TheCodexMe/
► TWITTER: https://twitter.com/thecodexme
► SUPPORT ME ON PATREON: https://www.patreon.com/The_Codex

♦♦♦♦♦♦♦♦♦♦

THECODEX

​A Practical Guide to Feature Engineering in Python

🔗 A Practical Guide to Feature Engineering in Python
Learn the underlying techniques and tools for effective feature engineering in Python

​CS221: Artificial Intelligence: Principles And Techniques | Stanford University

What do web search, speech recognition, face recognition, machine translation, autonomous driving, and automatic scheduling have in common? These are all complex real-world problems, and the goal of artificial intelligence (AI) is to tackle these with rigorous mathematical tools.

https://www.newworldai.com/cs221-artificial-intelligence-principles-and-techniques-stanford-university/

🔗 CS221: Artificial Intelligence: Principles and Techniques | Stanford University - New World : Artifi
CS221: Artificial Intelligence: Principles and Techniques | Stanford University

​🚶 HybridPose: 6D Object Pose Estimation under Hybrid Representations

https://github.com/chensong1995/HybridPose

Paper: https://arxiv.org/abs/2001.01869

Pixel-wise Voting Network for 6DoF Pose Estimation: https://github.com/zju3dv/pvnet

Наш телеграм канал - tglink.me/ai_machinelearning_big_data

🔗 chensong1995/HybridPose
Implementation of HybridPose: 6D Object Pose Estimation under Hybrid Representation - chensong1995/HybridPose

​Визуализация границ решения классификатора на основе изображений
Введение
Понимание того, как классификатор разбивает исходное многомерное пространство признаков на множество целевых классов, является важным шагом для анализа любой задачи классификации и оценки решения, полученного с помощью машинного обучения.
Современные подходы к визуализации решений классификаторов в основном либо используют диаграммы рассеивания, которые могут отображать лишь проекции исходных обучающих выборок, но явно не показывают фактические границы принятия решений, либо используют внутреннее устройство классификатора (например kNN, SVM, Logistic Regression) для которых легко построить геометрическую интерпретацию. Такой способ не подойдет для визуализации, например, нейросетевого классификатора.
В статье "Image-based Visualization of Classifier Decision Boundaries" (Rodrigues et al., 2018) предлагается эффективный, красивый и достаточно простой альтернативный метод для визуализации решений классификатора, который лишен вышеописанных недостатков. А именно метод подходит для классификаторов любого вида и строит изображение границы принятия решений с помощью изображений с произвольной частотой дискретизации.
Этот пост — краткий обзор основных идей и результатов из оригинальной статьи.

🔗 Визуализация границ решения классификатора на основе изображений
Введение Понимание того, как классификатор разбивает исходное многомерное пространство признаков на множество целевых классов, является важным шагом для анализа...

​Jupyter Notebooks 📓 from SIGMA Rules 🛡⚔️ to Query Elasticsearch 🏹

🔗 Jupyter Notebooks 📓 from SIGMA Rules 🛡⚔️ to Query Elasticsearch 🏹
Happy new year everyone 🎊! I’m taking a few days off before getting back to work and you know what that means 😆 Besides working out a…

​Neural Ordinary Differential Equations for Semantic Segmentation of Individual Colon Glands
Hans Pinckaers, Geert Litjens : https://arxiv.org/abs/1910.10470
GitHub : https://github.com/DIAGNijmegen/neural-odes-segmentation
#MedNeurIPS #NeurIPS #NeurIPS2019

🔗 Neural Ordinary Differential Equations for Semantic Segmentation of Individual Colon Glands
Automated medical image segmentation plays a key role in quantitative research and diagnostics. Convolutional neural networks based on the U-Net architecture are the state-of-the-art. A key disadvantage is the hard-coding of the receptive field size, which requires architecture optimization for each segmentation task. Furthermore, increasing the receptive field results in an increasing number of weights. Recently, Neural Ordinary Differential Equations (NODE) have been proposed, a new type of continuous depth deep neural network. This framework allows for a dynamic receptive field at a fixed memory cost and a smaller amount of parameters. We show on a colon gland segmentation dataset (GlaS) that these NODEs can be used within the U-Net framework to improve segmentation results while reducing memory load and parameter counts.

​Class-balanced Grouping and Sampling for Point Cloud 3D Object Detection

https://github.com/poodarchu/Det3D

https://arxiv.org/abs/1908.09492v1

🔗 poodarchu/Det3D
A general 3D object detection codebse. Contribute to poodarchu/Det3D development by creating an account on GitHub.