Automatically download and crop key information from the arxiv daily paper. (cpu version)

Related tags

Downloader FocusAX
Overview

FocusAX

按关键词筛选arxiv每日最新paper或从arxiv搜索。

  • 自动下载、获取摘要、自动截取文中表格和图片。

安装必要的环境

  • 安装 paddle
# GPU安装
python3 -m pip install paddlepaddle-gpu==2.1.1 -i https://mirror.baidu.com/pypi/simple

# CPU安装
 python3 -m pip install paddlepaddle==2.1.1 -i https://mirror.baidu.com/pypi/simple
  • 安装 Layout-Parser
=2.2"">
pip3 install -U https://paddleocr.bj.bcebos.com/whl/layoutparser-0.0.0-py3-none-any.whl
pip install "paddleocr>=2.2"
  • 按照其他必要的包
pip3 install -r requirements.txt
  • 下载模型权重
  • PubLayNet 下载解压后放置在paperparse目录下。目录结构如下
FocusAX
    - paperparse
        - ppyolov2_r50vd_dcn_365e_publaynet
            - inference.pdiparams
            - inference.pdiparams.info
            - inference.pdmodel
        - ...
    - downloader
        - ...
    - utils
        - ...
    - configs.py
    - focus_daily.py
    - focus_search.py
    - README.py
    - ...

使用教程

  • configs.py :程序参数配置文件
# =============== 网络代理 ================
# proxy = None # 不使用代理
proxy = {"http": "socks5://127.0.0.1:8080", "https": "socks5://127.0.0.1:8080"}
# =============== 保存文件根目录 ================
root_path = "./arxiv"
# =============== DNN模型推理配置信息 ================
threshold = 0.5
enable_mkldnn = True
enforce_cpu = True
thread_num = 4
  • focus_daily.py :按关键字过滤arxiv daily上的文章(仅当日)
if __name__ == '__main__':
    key_words = ['GAN'] # 要包含的关键词
    subject_words = ['ML', 'CV', 'AI']  # 要包含的类别
    start_parse(key_words, subject_words, needPDF=True, needZip=False)
  • focus_search.py :按关键字在arxiv检索
start_parse('Keyword')
  • root_path 目录中将创建新的文件夹保存结果

效果图

每个文件夹中的abs.md文件保留的是当前pdf的介绍,使用Typora等markdown编辑器打开。 image

image

ps:论文排版不规范会导致截图混乱。

其他

You might also like...
FireDM is a python open source (Internet Download Manager) with multi-connections, high speed engine, it downloads general files and videos from youtube and tons of other streaming websites .
FireDM is a python open source (Internet Download Manager) with multi-connections, high speed engine, it downloads general files and videos from youtube and tons of other streaming websites .

python open source (Internet Download Manager) with multi-connections, high speed engine, based on python, LibCurl, and youtube_dl https://github.com/firedm/FireDM

Using Youtube downloader is the fast and easy way to download and save any YouTube video.
Using Youtube downloader is the fast and easy way to download and save any YouTube video.

Youtube video downloader using Django Using Django as a backend along with pytube module to create Youtbue Video Downloader. https://yt-videos-downloa

Advance Image Downloader/Extractor (Job) is a Python-Flask web-based app, which will help the user download the any kind of Images at any date and time over the internet. These images will get downloaded as a job and then let user know that the images have been downloaded by sending them a link over an email.
Download and save Bing wallpapers and set as background for GNOME desktop

Save Bing wallpapers and set as background for GNOME desktop This script downloads the Bing wallpaper and sets it in the background of your gnome desk

Command-line program to download videos from YouTube.com and other video sites

youtube-dl - download videos from youtube.com or other video platforms

The free and open-source Download Manager written in pure Python
The free and open-source Download Manager written in pure Python

The free and open-source Download Manager written in pure Python

A Udemy downloader that can download DRM protected videos and non-DRM protected videos.
A Udemy downloader that can download DRM protected videos and non-DRM protected videos.

Udemy Downloader with DRM support NOTE This program is WIP, the code is provided as-is and i am not held resposible for any legal repercussions result

Scripts to download files and folders programmatically from Google Drive

Google Drive Downloader Scripts Every time I need to download a lot of files from Google Drive (e.g. a dataset), it's always incredibly frustrating an

Owner
HeoLis
Interesting in generate methods.
HeoLis
A toolkit to automatically crawl the paper list and download paper pdfs of ACL Ahthology.

ACL-Anthology-Crawler A toolkit to automatically crawl the paper list and download paper pdfs of ACL Anthology

Ray GG 9 Oct 9, 2022
Download Apple Music Cover Artwork in the best Quality by providing an Apple Music Link. It downloads the jpg, png and webp version since they often differ from another.

amogus.py - Version 0.0.5 amogus - Apple Music Hi-Res Artwork Fetcher this is my first real python tool so sorry if its bad amogus is a Python script

reaper 46 Jan 9, 2023
A modern CLI to download animes automatically from Twist

Kurby Kurby is a nice and simple CLI that use Twist website, and their huge collection to download animes for free and automatically Animes from Twist

Alain BERRIER 48 Dec 22, 2022
Automatically download multiple papers by keywords in CVPR

Automatically download multiple papers by keywords in CVPR

null 46 Jun 8, 2022
Can automatically download mods from a Curseforge modpack

Curseforge-Modpack-Downloader A Python script which automatically downloads mods from a Curseforge modpack. Installing Dependencies ⚠ Make sure you ha

Rayr 1 Sep 20, 2022
A Celery application to collect data, download media and extract information from social media APIs

Project IBEX A Celery application to collect data, download media and extract information from social media APIs. Requirements You must have a Redis D

ibex 4 Dec 15, 2022
A tool to download program information from Bugcrowd, for use by researchers to compare programs they are eligible to participate in

Description bcstats is a tool which allows Bugcrowd researchers to download information about all accessible programs (public and private) into a sing

null 19 Oct 13, 2022
A simple python script to fetch Bing daily images and set them randomly using hsetroot

Daily bing walls A script to set the daily bing image as the wallpaper. It can also set from a random set of previously fetched wallpapers. usage: dai

Harinarayanan K S 5 Nov 23, 2022
Arxiv2Kindle is a simple script written in python that converts LaTeX source downloaded from Arxiv and recompiles it to better fit a Kindle or other similar reading devices.

Arxiv2Kindle is a simple script written in python that converts LaTeX source downloaded from Arxiv and recompiles it to better fit a read

Soumik Rakshit 8 Jul 9, 2022
DYA ( Ditch YouTube API ) is a package created to power the user with YouTube Data API functionality without any API Key

Ditch YouTubeAPI (BETA) DYA ( Ditch YouTube API ) is a package created to power the user with YouTube Data API functionality without any API Key Detai

Sougata Jana 23 Dec 22, 2022