Are you looking for an efficient and modern framework to create your deep learning model? Look no further than PyTorch!
https://blog.paperspace.com/why-use-pytorch-deep-learning-framework/

YOLO v4 released! 🎉🎉
YOLO Is Back! Version 4 Boasts Improved Speed and Accuracy

The introduction of state-of-the-art, real-time object detection system YOLO (You only look once) in 2016 was a milestone in object detection research and led to better, faster and more accurate computer vision algorithms. Unfortunately, two months ago the father of YOLO Joseph Redmon announced he was leaving the field of computer vision due to concerns regarding the possible negative impact of his work. Redmon’s withdrawal triggered online debates and raised an important question: would there still be YOLO updates in the future?

Well, to the relief of many in the computer vision (CV) community, the answer is yes! The official YOLO Github account released an updated YOLO Version 4 last Friday.

YOLO v4 release lists three authors: Alexey Bochkovskiy, the Russian developer who built the YOLO Windows version, Chien-Yao Wang, and Hong-Yuan Mark Liao.

Compared with the previous YOLOv3, YOLOv4 has the following advantages:

👉 It is an efficient and powerful object detection model that enables anyone with a 1080 Ti or 2080 Ti GPU to train a super fast and accurate object detector.
👉 The influence of state-of-the-art “Bag-of-Freebies” and “Bag-of-Specials” object detection methods during detector training has been verified.
👉 The modified state-of-the-art methods, including CBN (Cross-iteration batch normalization), PAN (Path aggregation network), etc., are now more efficient and suitable for single GPU training.

The authors used s combined the following new features to make their design suitable for efficient training and detection:

👉 Weighted-Residual-Connections (WRC)
👉 Cross-Stage-Partial-Connections (CSP),
👉 A new backbone that can enhance learning capability of CNN.
👉 Cross mini-Batch Normalization (CmBN), represents a CBN modified version which assumes a batch contains four mini-batches
👉 Self-adversarial-training (SAT), represents a new data augmentation technique that operates in 2 forward backward stages
👉 Mish-activation, A novel self regularized non-monotonic neural activation function
👉 Mosaic data augmentation, represents a new data augmentation method that mixes 4 training images instead of a single image.
👉 DropBlock regularization, a better regularization method for CNN
👉 CIoU loss, achieves better convergence speed and accuracy on the BBox regression problem.
YOLOv4’s excellent speed and accuracy and the well-written paper are a great contribution to engineering and academics. The update also illustrates an encouraging promotion and development of open source software: even if the father of YOLO has abandoned model updates, others can maintain and continue to promote the development of the powerful tools which we are increasingly reliant on.
The source code is on the Project Github. https://github.com/AlexeyAB/darknet
The paper YOLOv4: Optimal Speed and Accuracy of Object Detection is on arxiv.

People like to use cool names which are often confusing. When I started playing with CNN beyond single label classification, I got confused with the different names and formulations people write in their papers, and even with the loss layer names of the deep learning frameworks such as Caffe, Pytorch or TensorFlow. This blog groups up the different names and variations people use for Cross-Entropy Loss
https://gombru.github.io/2018/05/23/cross_entropy_loss/

Live interactive CNN tutorial running in your browser! Using this tool, you can visualize your CNN, interact with individual components, and intermediate layers. One of the best CNN refresher out there! Check it out using the following link. https://poloclub.github.io/cnn-explainer/

Google AI Team Open Sources BiT - Big Transfer: General Visual Representation Learning (Computer Vision)
Paper: https://arxiv.org/abs/1912.11370
Github: https://github.com/google-research/big_transfer
https://ai.googleblog.com/2020/05/open-sourcing-bit-exploring-large-scale.html

All Coursera Courses for FREE 🥳🥳

Learning with Coursera for college, University students
Starting today, college and university students around the world can learn and earn certificates on Coursera for free. Current undergraduate, graduate, or recently graduated students with a verified school email can sign up to get free access to over 3,800 courses, 150 Guided Projects, 400 Specializations, and 11 Professional Certificates. They can enroll in programs for free until July 31 — no credit card required. Once enrolled, they will have until Sept. 30, 2020, to complete the programs.
https://www.coursera.org/for-university-and-college-students

Drive-Net: Convolutional Network for Driver Distraction Detection
https://arxiv.org/abs/2006.12586

Hi
I am planning to share my startup experience and learning in this Telegram group. @StartupMakeinIndia

"Hellooo World!"

Our next virtual meetup is here!! 💫

OpenCV: The Untold Story with Satya Mallick

Ai India would like to announce that our second virtual AI meetup with Satya Mallick (CEO, OpenCV.org)

About speaker:
Dr. Satya Mallick is the CEO of OpenCV. In 2017, IBM Watson’s blog named Dr. Mallick as one of the top 30 AI influencers to follow on Twitter. His work has been featured in publications like the BBC, Time, Huffington Post, the Wall Street Journal, Oprah Magazine, TechCrunch, and Theregister UK.

Join us for the second biggest AI Meet-up to know

To Know How OpenCV was Born?
To Know how it changed the world of Computer Vision?
To know how it works ?
To know how learn OpenCV and become computer vision expert?
To know how OpenCV AI Kit is gonna fight with Raspberry PI?
To know the about Kickstarter campaign for the OpenCV AI Kit

Date : 9th August 2020
Time : 7.15 pm (IST)

Register for FREE : https://bit.ly/aiindiawithopencv

Come to learn OpenCV at virtual meetup, we'd love to see you!

Facebook Research recently open-sourced their new library - TransCoder.

TransCoder is a model which uses unsupervised deep-learning which can translate code from Python to C++ & it outperforms rule-based translation programs.
https://towardsdatascience.com/facebooks-transcoder-an-ai-source-to-source-compiler-23ea77f3234b

Deep Learning with PyTorch
Download a free copy of the full book and learn how to get started with AI / ML development using PyTorch https://pytorch.org/deep-learning-with-pytorch

The Deep Learning Lecture Series

The Deep Learning Lecture Series is a collaboration between DeepMind and the UCL Centre for Artificial Intelligence.
https://www.youtube.com/playlist?list=PLqYmG7hTraZCDxZ44o4p3N5Anz3lLRVZF&fbclid=IwAR0yY9iPdEB6-dIGkPg3C70aFCeVLK9bC4G7Zuj_qBgv5s42DLBrOPCWJgU