Data Science | Machine Learning with Python for Researchers
31.5K subscribers
1.56K photos
102 videos
22 files
1.83K links
Admin: @HusseinSheikho

The Data Science and Python channel is for researchers and advanced programmers

Buy ads: https://telega.io/c/dataScienceT
Download Telegram
๐Ÿš€ NAUTILUS: boosting Bayesian importance nested sampling with deep learning

A novel approach to boost the efficiency of the importance nested sampling (INS) technique for Bayesian posterior and evidence estimation using deep learning.

Install:
pip install nautilus-sampler

import corner
import numpy as np
from nautilus import Prior, Sampler
from scipy.stats import multivariate_normal

prior = Prior()
for key in 'abc':
prior.add_parameter(key)

def likelihood(param_dict):
x = [param_dict[key] for key in 'abc']
return multivariate_normal.logpdf(x, mean=[0.4, 0.5, 0.6], cov=0.01)

sampler = Sampler(prior, likelihood)
sampler.run(verbose=True)
points, log_w, log_l = sampler.posterior()
corner.corner(points, weights=np.exp(log_w), labels='abc')

๐Ÿ–ฅ Github: https://github.com/johannesulf/nautilus

โญ๏ธ Docs: https://nautilus-sampler.readthedocs.io/

๐Ÿ“• Paper: https://arxiv.org/abs/2306.16923v1

https://t.iss.one/DataScienceT
โค6
๐ŸŒ๏ธ GlOttal-flow LPC Filter (GOLF)

A DDSP-based neural vocoder.

๐Ÿ–ฅ Github: https://github.com/yoyololicon/golf

๐Ÿ“• Paper: https://arxiv.org/abs/2306.17252v1

๐Ÿ”—Demo: https://yoyololicon.github.io/golf-demo/

https://t.iss.one/DataScienceT
โคโ€๐Ÿ”ฅ3โค1๐Ÿ‘1
This media is not supported in your browser
VIEW IN TELEGRAM
๐Ÿ”ฎ SAM-PT: Segment Anything + Tracking ๐Ÿ”ฎ

โญ๏ธ SAM-PT is the first method to utilize sparse point propagation for Video Object Segmentation (VOS).

๐ŸŒ Review https://t.ly/QLMG

๐ŸŒ Paper arxiv.org/pdf/2307.01197.pdf

๐ŸŒ Project www.vis.xyz/pub/sam-pt/

๐ŸŒ Code github.com/SysCV/sam-pt

https://t.iss.one/DataScienceT
โคโ€๐Ÿ”ฅ1โค1๐Ÿ‘1
๐ŸธThe Drunkardโ€™s Odometry: Estimating Camera Motion in Deforming Scenes

๐Ÿ–ฅ Github: https://github.com/UZ-SLAMLab/DrunkardsOdometry

โฉ Paper: https://arxiv.org/pdf/2306.16917v1.pdf

๐Ÿ’จ Dataset: https://paperswithcode.com/dataset/drunkard-s-dataset

https://t.iss.one/DataScienceT
โคโ€๐Ÿ”ฅ2
โคโ€๐Ÿ”ฅ1โค1๐Ÿ‘1
This media is not supported in your browser
VIEW IN TELEGRAM
๐Ÿชฉ DISCO: Human Dance Generation

โญ๏ธ NTU (+ #Microsoft) unveils DISCO: a big step towards the Human Dance Generation.

๐ŸŒ Review https://t.ly/cNGX

๐ŸŒ Paper arxiv.org/pdf/2307.00040.pdf

๐ŸŒProject: disco-dance.github.io/

๐ŸŒ Code github.com/Wangt-CN/DisCo

https://t.iss.one/DataScienceT
๐Ÿ‘3โค1
Building an Image Recognition API using Flask.

Step 1: Set up the project environment

1. Create a new directory for your project and navigate to it.
2. Create a virtual environment (optional but recommended):
(Image 1.)
3. Install the necessary libraries (image 2.)

Step 2: Create a Flask Web Application
Create a new file called app.py in the project directory (image 3.)

Step 3: Launch the Flask Application
Save the changes and run the Flask application (image 4.)

Step 4: Test the API
Your API is now up and running and you can send images to /predict via HTTP POST requests.
You can use tools such as curl or Postman to test the API.
โ€ข An example of using curl (image 5.)
โ€ข An example using Python queries (image 6.)

https://t.iss.one/DataScienceT
โคโ€๐Ÿ”ฅ2๐Ÿ˜2๐Ÿ‘1
Deep Learning
NLP
AI
Python
ML
Data Mining
Tensorflow
Keras

๐Ÿ‘‡๐Ÿ‘‡๐Ÿ‘‡๐Ÿ‘‡๐Ÿ‘‡
@Machine_learn
โคโ€๐Ÿ”ฅ3โค2๐Ÿ˜1
This media is not supported in your browser
VIEW IN TELEGRAM
๐Ÿš€ Hierarchical Open-vocabulary Universal Image Segmentation

Decoupled text-image fusion mechanism and representation learning modules for both "things" and "stuff".

๐Ÿ–ฅ Github: https://github.com/berkeley-hipie/hipie

๐Ÿ“• Paper: https://arxiv.org/abs/2307.00764v1

๐Ÿ”—Project: https://people.eecs.berkeley.edu/~xdwang/projects/HIPIE/

๐Ÿ”— Dataset: https://paperswithcode.com/dataset/pascal-panoptic-parts

https://t.iss.one/DataScienceT
โคโ€๐Ÿ”ฅ1
๐Ÿ”Foundation Model for Endoscopy Video Analysis

๐Ÿ–ฅ Github: https://github.com/med-air/endo-fm

โฉ Paper: https://arxiv.org/pdf/2306.16741v1.pdf

๐Ÿ’จ Dataset: https://paperswithcode.com/dataset/kumc

https://t.iss.one/DataScienceT
โคโ€๐Ÿ”ฅ3
We launched a special bot some time ago to download all scientific, software and mathematics books The bot contains more than thirty million books, and new books are downloaded first, In addition to the possibility of downloading all articles and scientific papers for free

To request a subscription: talk to @Hussein_Sheikho
โค3๐Ÿ‘3
This media is not supported in your browser
VIEW IN TELEGRAM
๐ŸŽจ Making ML-powered web games with Transformers.js

The goal of this tutorial is to show you how easy it is to make your own ML-powered web game.

๐Ÿ–ฅ Github: https://github.com/xenova/doodle-dash

๐Ÿค— Hugging face: https://huggingface.co/blog/ml-web-games

โญ๏ธ Code: https://github.com/xenova/doodle-dash

๐Ÿ”—Demo: https://huggingface.co/spaces/Xenova/doodle-dash

๐Ÿ”— Dataset: https://huggingface.co/datasets/Xenova/quickdraw-small

https://t.iss.one/DataScienceT
โค1๐Ÿ‘1
๐Ÿฆ™ Focused Transformer: Contrastive Training for Context Scaling

LongLLaMA, a large language model capable of handling long contexts of 256k tokens or even more.

๐Ÿ–ฅ Github: https://github.com/cstankonrad/long_llama

๐Ÿ“• Paper: https://arxiv.org/abs/2307.03170v1

๐Ÿ–ฅ Colab: https://colab.research.google.com/github/CStanKonrad/long_llama/blob/main/long_llama_colab.ipynb

๐Ÿ”— Dataset: https://paperswithcode.com/dataset/pg-19

https://t.iss.one/DataScienceT
๐Ÿ‘3โค1โคโ€๐Ÿ”ฅ1
This media is not supported in your browser
VIEW IN TELEGRAM
ContainerGym: A Real-World Reinforcement Learning Benchmark for Resource Allocation โ™ป๏ธ

๐Ÿ–ฅ Github: https://github.com/pendu/containergym

โฉ Paper: https://arxiv.org/pdf/2307.02991v1.pdf

๐Ÿ’จ Dataset: https://paperswithcode.com/dataset/openai-gym

https://t.iss.one/DataScienceT
๐Ÿ‘1