Pro-Logic

09/10/2019

09/10/2019

Pythia : New deep learning framework from Facebook

Pythia is a deep learning framework that supports multitasking in the vision and language domain. Built on our open-source framework, the modular, plug-and-play design enables researchers to quickly build, reproduce, and benchmark AI models. is designed for vision and language tasks, such as answering questions related to visual data and automatically generating image caption

Link: https://engineering.fb.com/ai-research/pythia/
GitHub: https://github.com/facebookresearch/pythia
more : https://arxiv.org/abs/1807.09956

Pythia is a new open source deep learning framework that enables researchers to quickly build, reproduce, and benchmark AI models.

09/10/2019

Awesome news for all NLP people 🔥! PyTorch-Transformers 1.0 is out (formerly known as pytorch-pretrained-bert). It now includes 6 SOTA NLP models: Bert, GPT, GPT-2, Transformer-XL, XLNet, XLM and also 27 pre-trained weights. There's also a new unified API for models and tokenizer. Check the release note for more information and also the project in general. Great work by the team at Hugging Face. Can't wait to use it for our own projects.

🔤 Code: https://github.com/huggingface/transformers
📘 Documentation: https://huggingface.co/transformers/
📝 Release note: https://github.com/huggingface/transformers/releases/tag/v1.0.0

🤗 Transformers: State-of-the-art Natural Language Processing for TensorFlow 2.0 and PyTorch. - huggingface/transformers

09/10/2019

FSGAN: Subject Agnostic Face Swapping and Reenactment

New paper on creation

YouTube demo: https://www.youtube.com/watch?v=duo-tHbSdMk
Link: https://nirkin.com/fsgan/
ArXiV:https://arxiv.org/pdf/1908.05932.pdf

📝 The paper "FSGAN: Subject Agnostic Face Swapping and Reenactment" is available here: https://nirkin.com/fsgan/ ❤️ Pick up cool perks on our Patreon page: h...

09/10/2019

Great collections of Data Science learning materials

The list includes free books and online courses on range of DS-related disciplines

Notable researchers, papers and even datasets. It is a great place to start reviewing your knowledge or learning something new.

Link: https://hackmd.io/VASaLIb1Quunia9Nva7gpA

09/10/2019

TensorFlow 2.0.0

Release 2.0.0 - Major Features and Improvements

https://github.com/tensorflow/tensorflow/releases

https://medium.com/tensorflow/tensorflow-2-0-is-now-available-57d706c2a9ab

TensorFlow 2.0 focuses on simplicity and ease of use, featuring updates like:

- Easy model building with Keras and eager ex*****on.
- Robust model deployment in production on any platform.
- Powerful experimentation for research.
- API simplification by reducing duplication and removing deprecated endpoints.

For details on best practices with 2.0, see the Effective 2.0 guide - https://www.tensorflow.org/guide/effective_tf2

An Open Source Machine Learning Framework for Everyone - tensorflow/tensorflow

09/10/2019

A Report by H20

In this report you’ll find:

https://drive.google.com/open?id=1PQVYoNMFjWvNOc7OGsSAjgeVTby2ImKI

Definitions and examples
Social and Commercial Motivations for Machine Learning
A Machine Learning Interpretability Taxonomy for Applied Practitioners
Common Interpretability Techniques
Limitations and Precautions
Testing Interpretability and Fairness
Machine Learning Interpretability in Action

09/10/2019

A step-by-step guide on how to learn Computer Vision, Deep Learning, and OpenCV by pyimagesearch - Adrian Rosebrock

https://www.pyimagesearch.com/start-here/

Inside the guide you'll find step-by-step, easy to follow instructions on:

Getting started with OpenCV
Deep Learning (Keras and TensorFlow)
Face applications (including face detection, facial landmarks, and face recognition)
Optical Character Recognition (OCR)
Object detection (SSDs, YOLO, etc.)
Object tracking
Instance segmentation (Mask R-CNNs)
CV and DL on embedded/IoT devices (RPi, Movidius NCS Google Coral, Jetson Nano)
Medical Computer Vision..and much more!