GraphParsingTool

Graph Parsing Tool for CROCS 2024

usage

Tool Name: ssml_parser.py

usage: [-h] [-input INPUT] [-img_prefix IMG_PREFIX] [-output OUTPUT] [-num_workers NUM_WORKERS]

optional arguments:
 -h, --help            show this help message and exit
 -input INPUT
 -img_prefix IMG_PREFIX
 -output OUTPUT

input: the lab file or result file. Details of file format can be seen in CROCS 2024 Homepage(https://crocs-ifly-ustc.github.io/crocs/data.html)

img_prefix: the directory of all images

output: the output directory. For each sample, the tool will save a jpeg file and a json file with same name to this directory. The jpeg file contains the raw image and visual results of chemical structures in its label or result string. The json file records the graph data of all chemical structures parsed from label or result string. See train_48095.jpg and train_48095.json for example.