getpaper 0.1.0

getpaper - papers download made easy!

getpaper

Paper downloader

getting started

Install the library with:

pip install getpaper

Usage

Downloading papers

After the installation you can either import the library into your python code or you can use the console scripts, for example:

download download download_pubmed --pubmed 22266545 --folder papers --name pmid

Downloads the paper with pubmed id into the folder 'papers' and uses the pubmed id as name

download download download_doi --doi 10.1519/JSC.0b013e318225bbae --folder papers

Downloads the paper with DOI into the folder papers, as --name is not specified doi is used as name

Parsing the papers

You can parse the downloaded papers with the unstructure library. For example if the papers are in the folder test, you can run:

getpaper/parse.py parse_folder --folder /home/antonkulaga/sources/getpaper/test

You can also parse papers on a per file basis, for example:

getpaper/parse.py parse_paper --paper /home/antonkulaga/sources/getpaper/test/22266545.pdf

Indexing papers

We also provide features to index the papers with openai or lambda embeddings and save them in chromadb vector store. For openai embeddings to work you have to create .env file and specify your openai key there, see .env.template as example For example if you have your papers inside data/output/test/papers folder and you want to make a ChromaDB index at data/output/test/index you can do it by:

python getpaper/index.py index_papers --papers data/output/test/papers --folder data/output/test/index --collection mypapers --chunk_size 6000

Examples

You can run examples.py to see usage examples

Additional requirements

Detectron2 is required for using models from the layoutparser model zoo but is not automatically installed with this package. For MacOS and Linux, build from source with:

pip install 'git+https://github.com/facebookresearch/detectron2.git@e2ce8dc#egg=detectron2'