ds3lab / docparser Goto Github PK
View Code? Open in Web Editor NEWLicense: MIT License
License: MIT License
When will the code be released?
I want to ask if I want to use this model to classify some new pages, where should I put my data?
Is there an inference code to pass my own input document using your pre-trained model and get the detected tables output?
Thank you.
Hello, thank you for the effort you've put into this repository. I've recently tried to run it against my own pdfs (converted to images)
And this was result:
My code to draw:
import argparse
import json
import logging
import os
import cv2
import skimage.io
from docparser import stage1_entity_detector
image_name="outputname-01.png"
sample_img_path = "/home/ubuntu/projects/pdf-expirements/DocParser/images/{}".format(image_name)
output_dir = "/home/ubuntu/projects/pdf-expirements/DocParser/demos/output"
entity_detector = stage1_entity_detector.EntityDetector()
entity_detector.init_model(default_weights="highlevel_wsft")
sample_img = skimage.io.imread(sample_img_path)
results = entity_detector.predict(sample_img)
image = cv2.imread(sample_img_path)
shapes = []
for pred in results["prediction_list"]:
box = pred["bbox_orig_coords"].tolist()
label = pred["class_name"]
score = pred["pred_score"]
shapes.append({"box": box, "label": label, "score": score})
for shape in shapes:
y1, x1, y2, x2 = shape["box"]
image = cv2.rectangle(image, (x1, y1), (x2, y2), (255, 0, 0), 2)
cv2.imwrite("{}/{}".format(output_dir, image_name), image)
Most of the bounding boxes were drawn very incorrectly. Am I doing the drawing wrong, or does it not work with files, which contain a structure like this. ? Thank you for your time.
Hello,
Thanks for your work. I am trying to use your model to generate structures for the tables in ICDAR 2019 TRACK B2 Modern images. Please can you point me to the file to use and how to generate the structures for the tables in those images?
Thanks
Hi,
Thanks for sharing the DocParser!
Is there a license associated with this repo?
Thanks in advance!
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.