Catégorie : Machine Learning
-
GitHub – renatoviolin/Multiple-Choice-Question-Generation-T5-and-Text2Text: Question Generation using Google T5 and Text2Text
https://github.com/renatoviolin/Multiple-Choice-Question-Generation-T5-and-Text2Text
-
GitHub – AkariAsai/learning_to_retrieve_reasoning_paths: The official implementation of ICLR 2020, « Learning to Retrieve Reasoning Paths over Wikipedia Graph for Question Answering ».
https://github.com/AkariAsai/learning_to_retrieve_reasoning_paths
-
turbo_transformers: a fast and user-friendly runtime for transformer inference on CPU and GPU
Transformer is the most critical alogrithm innovation in the NLP field in recent years. It brings higher model accuracy while introduces more calculations. The efficient deployment of online Transformer-based services faces enormous challenges. In order to make the costly Transformer online service more efficient, the WeChat AI open-sourced a Transformer inference acceleration tool called TurboTransformers,…
-
Differentiable SDE solvers with GPU support and efficient sensitivity analysis.
https://github.com/google-research/torchsde GitHub – google-research/torchsde: Differentiable SDE solvers with GPU support and efficient sensitivity analysis. Examples. demo.ipynb in the examples folder is a short guide on how one may use the codebase for solving SDEs without considering gradient computation. It covers subtle points such as fixing the randomness in the solver and the consequence of noise…
-
GitHub – graphbrain/graphbrain: Language, Knowledge, Cognition
https://github.com/graphbrain/graphbrain
-
Using GitHub Actions for MLOps & Data Science
https://github.blog/2020-06-17-using-github-actions-for-mlops-data-science/ Using GitHub Actions for MLOps & Data Science – The GitHub Blog Background. Machine Learning Operations (or MLOps) enables Data Scientists to work in a more collaborative fashion, by providing testing, lineage, versioning, and historical information in an automated way. Because the landscape of MLOps is nascent, data scientists are often forced to implement…
-
What I learned from looking at 200 machine learning tools
https://huyenchip.com/2020/06/22/mlops.html
-
Bert: Step by step by Hugging face – The Startup – Medium
https://medium.com/swlh/bert-step-by-step-b7ff47fcfbe
-
GitHub – huggingface/nlp: 🤗 nlp: datasets and evaluation metrics for Natural Language P rocessing in NumPy, Pandas, PyTorch and TensorFlow
https://github.com/huggingface/nlp
-
Microsoft’s OpenAI supercomputer has 285,000 CPU cores, 10,000 GPUs | Engadget
https://www-engadget-com.cdn.ampproject.org/c/s/www.engadget.com/amp/microsoft-openai-supercomputer-azure-150001119.html
-
A Step by Step Guide to Tracking Hugging Face Model Performance | huggingface-demo | W&B
https://app.wandb.ai/jxmorris12/huggingface-demo/reports/A-Step-by-Step-Guide-to-Tracking-Hugging-Face-Model-Performance–VmlldzoxMDE2MTU
-
Introducing PyTorch BigGraph – Towards Data Science
https://towardsdatascience.com/introducing-pytorch-biggraph-4aeaec1559ec
-
Working with Hugging Face Transformers and TF 2.0 – Towards Data Science
https://towardsdatascience.com/working-with-hugging-face-transformers-and-tf-2-0-89bf35e3555a
-
Transfer Learning on HuggingFace BERT – Sentence correctness classification
https://www.youtube.com/watch?v=WiHvs1PmZ_4&feature=youtu.be
-
GitHub – huggingface/awesome-papers: Papers & presentations from Hugging Face’s weekly science day
https://github.com/huggingface/awesome-papers
-
MLOps feature dive: Manage your assets, artifacts and code | AI Show | Channel 9
https://channel9.msdn.com/Shows/AI-Show/MLOps-feature-dive-Manage-your-assets-artifacts-and-code
-
Approaching (almost) Any NLP Problem
https://www.slideshare.net/mobile/abhishekkrthakur/approaching-almost-any-nlp-problem
-
2001.09977 Towards a Human-like Open-Domain Chatbot
https://arxiv.org/abs/2001.09977
-
Satoshi Iizuka — DeepRemaster
http://iizuka.cs.tsukuba.ac.jp/projects/remastering/en/index.html
-
Top 4 libraries you must know before diving into any deep learning projects
https://medium.com/@abhisheksingh007226/top-4-libraries-you-must-know-before-diving-into-any-deep-learning-projects-61904286478d Top 4 libraries you must know before diving into any deep learning projects Libraries play a very important role in solving any problem. It makes our task easier. For example we can wrap image classification task… medium.com
-
GitHub – skorch-dev/skorch: A scikit-learn compatible neural network library that wraps pytorch
https://github.com/skorch-dev/skorch GitHub – skorch-dev/skorch: A scikit-learn compatible neural network library that wraps pytorch * Some cleanups in test_scoring * indentation level * disable some pylint messages * unused fixtures * Don’t iterate over data when using cached scoring Before, net.infer was cached when using a scoring callback wiht use_caching=True. github.com
-
https://www.microsoft.com/en-us/research/blog/turing-nlg-a-17-billion-parameter-language-model-by-microsoft/
https://www.microsoft.com/en-us/research/blog/turing-nlg-a-17-billion-parameter-language-model-by-microsoft/
-
How many tech skills are enough for a data scientist?
https://towardsdatascience.com/most-essential-set-of-skills-for-the-type-of-data-scientist-job-youre-looking-for-44d5f6d23ca3
-
Production Machine Learning Pipeline for Text Classification with fastText
https://blog.valohai.com/production-machine-learning-pipeline-text-classification-fasttext
-
Monk_Object_Detection/Example – Indoor Image Object Detection and Tagging.ipynb at master · Tessellate-Imaging/Monk_Object_Detection · GitHub
Experimented with multi-gpu training of indoor object detector using RetinaNet and Open-Images – V5 dataset The detector consists of 24 classes such as table, bed, sofas, home and kitchen appliances, etc. The training ran on AWS P3.x large instances4 Nvidia V100 GPUS 244 GB CPU RAM 32 CPUs Training time – 5 hours for 10…
-
https://www.zdnet.com/google-amp/article/facebooks-latest-giant-language-ai-hits-computing-wall-at-500-nvidia-gpus/
https://www.zdnet.com/google-amp/article/facebooks-latest-giant-language-ai-hits-computing-wall-at-500-nvidia-gpus/
-
Hugging Face launches popular Transformers NLP library for TensorFlow | VentureBeat
https://venturebeat.com/2019/09/26/hugging-face-launches-popular-transformers-nlp-library-for-tensorflow/
-
The Best 4-GPU Deep Learning Rig only costs $7000 not $11,000.
https://l7.curtisnorthcutt.com/the-best-4-gpu-deep-learning-rig
-
oborchers/Fast_Sentence_Embeddings: Compute Sentence Embeddings Fast!
https://github.com/oborchers/Fast_Sentence_Embeddings/
-
Five Hypotheses as to why Artificial Intelligence and Machine Learning projects fail
https://towardsdatascience.com/five-hypotheses-as-to-why-artificial-intelligence-and-machine-learning-projects-fail-7c6b2c456d41
-
https://openai.com/blog/gpt-2-6-month-follow-up/
https://openai.com/blog/gpt-2-6-month-follow-up/
-
2 Getting started with ggplot2 | ggplot2: Elegant Graphics for Data Analysis
https://ggplot2-book.org/getting-started.html
-
Microsoft Machine Learning for Apache Spark
https://mmlspark.blob.core.windows.net/website/index.html?fbclid=IwAR34cUjp-DtjR9AspRiGYmrngkPJLRxMJ2U6afAUSEKgQBatDf6s7QjyJg0
-
Beat the Odds: How to Conquer Common AI Challenges – InformationWeek
https://www.informationweek.com/big-data/ai-machine-learning/beat-the-odds-how-to-conquer-common-ai-challenges/a/d-id/1335545
-
All-optical diffractive neural network closes performance gap with electronic neural networks
https://www.eurekalert.org/pub_releases/2019-08/ssfo-adn081219.php?fbclid=IwAR062lNMQY5Cmd380qwxZiX89ZfbJV6OWarYBnq8oVRYm80PZqjRFKqMe8Q
-
Google open-sources Live Transcribe’s speech engine | VentureBeat
https://venturebeat.com/2019/08/16/google-open-sources-live-transcribes-speech-engine/
-
12 NLP Researchers, Practitioners & Innovators You Should Be Following
https://www.kdnuggets.com/2019/08/nlp-researchers-practitioners-innovators-should-follow.html
-
Megatron-LM: Entering the Frontiers of NLP
https://blog.exxactcorp.com/megatron-lm-unleashed-nvidias-transformer-megatraon-lm-is-the-nlp-model-ever-trained/
-
Intel Introduces Pohoiki Beach, a New 64-Chip Neuromorphic System – News
https://www.allaboutcircuits.com/news/intel-introduces-pohoiki-beach-its-64-chip-neuromorphic-system/
-
https://www.profillic.com/paper/arxiv:1907.05337
https://www.profillic.com/paper/arxiv:1907.05337
-
Deep Forest: Towards An Alternative to Deep Neural Networks | IJCAI
https://www.ijcai.org/proceedings/2017/497
-
Profillic: AI research & source code to supercharge your projects
https://www.profillic.com/paper/arxiv:1906.01267
-
10 Best Data Science Reads for Students – Towards Data Science
https://towardsdatascience.com/10-best-data-science-reads-for-students-3bae97d9bb23
-
https://medium.com/deepquestai/deepstack-for-raspberry-pi-intel-neural-compute-acceleration-1cf7d0b5e859
https://medium.com/deepquestai/deepstack-for-raspberry-pi-intel-neural-compute-acceleration-1cf7d0b5e859
-
How to power up your product by machine learning with python microservice, pt. 1
https://medium.com/@dkisler/how-to-power-up-your-product-by-machine-learning-with-python-microservice-pt-1-de0f2b434bec
-
ndrplz/semiparametric: Semi-parametric Object Synthesis. A semi-parametric approach for synthesizing novel views of an object from a single monocular image.
https://github.com/ndrplz/semiparametric
-
DeepGraphLearning/LiteratureDL4Graph
https://github.com/DeepGraphLearning/LiteratureDL4Graph
-
A comprehensive guide to the state-of-art in how AI is transforming the visual effects (VFX) industry – Ross Dawson
https://rossdawson.com/futurist/implications-of-ai/comprehensive-guide-ai-artificial-intelligence-visual-effects-vfx/