Thomas Wolf’s post on how to make #Python #NLP faster.

Spoiler: use spaCy's internals and a bit of Cython magic

Bonus: a #Jupyter notebook with examples processing over 80 millions words per sec…

https://medium.com/huggingface/100-times-faster-natural-language-processing-in-python-ee32033bdced

ConvNet outperforms human dermatologists for melanoma detection.

Dermatologists in level-II protocol:
- sensitivity: 88.9% (±9.6%, P = 0.19)
- specificity: 75.7% (±11.7%, P < 0.05)

https://academic.oup.com/annonc/advance-article/doi/10.1093/annonc/mdy166/5004443

#cv #dl

Transfer Learning in Tensorflow (VGG19 on CIFAR-10): Part 1

Tutorial on how to use pretrained networks not to make them learn everything from scratch.

Link: https://towardsdatascience.com/transfer-learning-in-tensorflow-9e4f7eae3bb4

#tutorial #cv #tensorflow #dl

Introduction to numpy matrices syntax.

https://www.kdnuggets.com/2017/03/working-numpy-matrices.html

#numpy #Python #tutorial

Second part of the tutorial.

https://towardsdatascience.com/transfer-learning-in-tensorflow-5d2b6ad495cb

Brief intro into object detection with python.

https://towardsdatascience.com/object-detection-with-10-lines-of-code-d6cb4d86f606

#python #tensorflow #cv

New #facebook research paper on using #GAN for “opening” eyes on the portraits.

https://research.fb.com/publications/eye-in-painting-with-exemplar-generative-adversarial-networks/

Basic intro into #python #scraping with lxml and requests: https://dev.to/timber/an-intro-to-web-scraping-with-lxml-and-python-32hl

"Breast density classification with deep convolutional neural networks"

Github: https://github.com/nyukat/breast_density_classifier
Arxiv: https://arxiv.org/abs/1711.03674

By N. Wu, myself, Y. Shen, J. Su, S. G. Kim, E. Kim

Not quite machine learning, but close enough: method for calculating sqaure root inwardly.

DensePose: Dense Human Pose Estimation In The Wild

Facebook AI Research group presented a paper on pose estimation. That will help Facebook with better understanding of the processed videos.

NEW: DensePose-COCO, a large-scale ground-truth dataset with image-to-surface correspondences manually annotated on 50K COCO images.

Project website: http://densepose.org/
Arxiv: https://arxiv.org/abs/1802.00434

#facebook #fair #cvpr #cv #CNN #dataset

More papers are available at: https://research.fb.com/facebook-research-at-cvpr-2018/

Best paper award at #CVPR2018 main idea: study twenty five different visual tasks to understand how & when transfer learning works from one task to another, reducing demand for labelled data.

Video: https://youtu.be/SUq1CiX-KzM
Project website: http://taskonomy.stanford.edu
Arxiv: https://arxiv.org/pdf/1804.08328.pdf

How to Do Distributed Deep Learning for Object Detection Using Horovod on Azure

Link: https://blogs.technet.microsoft.com/machinelearning/2018/06/20/how-to-do-distributed-deep-learning-for-object-detection-using-horovod-on-azure/

#cloud #horovod #cv #dl

Another #CVPR2018 paper award: Global Pose Estimation with Attention-based RNNs.

Arxiv: https://arxiv.org/abs/1802.06857

#pose #rnn #dl

The natural language decathlon benchmark to do well on ten different nlp tasks including question answering, translation, summarization, sentiment analysis by Salesforce research team.

Project website: https://einstein.ai/research/the-natural-language-decathlon
PDF link: https://einstein.ai/static/images/pages/research/decaNLP/decaNLP.pdf
Github: https://github.com/salesforce/decaNLP

#nlp #dl

Microsoft Research announced Open Data project! This single, cloud-hosted location offers datasets representing many years of data curation and research efforts by Microsoft.

https://www.microsoft.com/en-us/research/blog/announcing-microsoft-research-open-data-datasets-by-microsoft-research-now-available-in-the-cloud/

#data #microsoft #opensource

Kaggle №1 Winning Approach for Plant Seedlings Classification


https://medium.com/@shridhar743/kaggle-1-winning-approach-for-image-classification-challenge-9c1188157a86

Results of the OpenAI's competition on reinforcement learning:
https://blog.openai.com/first-retro-contest-retrospective/

Hey guys (both male and female)! We are now 7157 and it’s time to find out a bit more about you.

Please, help the channel edition team to post more relivant information, by filling in the form:

https://goo.gl/forms/VBYApzVRGCUhzb713

A lil’ bit of interactive on @opendatascience: we updated the pre-last question with option «don’t remember».

And as a reminder: for those who want to suggest any news or posts, there is a special bot you can use to submit a post: @opendatasciencebot

80 responses / 1413 post views