DocClassifier
The core functionality of this project is called "Document Classification".
Upon seeing this title, you might smirk and think, "Isn't it just a classification model?"
- Yes, and no.
This time, we aim to create an atypical classification model. While its application scope may be limited, its intrinsic interest is quite high.
It might not be what you imagine. Please continue reading.
This project has a unique origin. It was conceived by my friend: Jack, L.
Because he didn't have time to write the web page, he completed the early-stage program development and feasibility verification first. Then, he entrusted me with this idea, allowing me to continue to complete the details and publish them here. I would like to express my special thanks to him for his contribution.
ποΈ Introduction
In past project experiences, the classification model can be considered one of the most common machine learning tasks.
ποΈ Installation
Currently, there is no installation package available on PyPI, and there are no plans for one in the near future.
ποΈ QuickStart
We provide a simple model inference interface that includes logic for preprocessing and postprocessing.
ποΈ Advanced
When invoking the DocClassifier model, you can perform advanced settings by passing parameters.
ποΈ Model Design
A comprehensive model functionality is not achieved overnight; it requires multiple iterations of adjustments and designs.
ποΈ Evaluation
The test dataset for this project is provided by a private institution. For privacy protection, we only provide the evaluation results of this dataset.
ποΈ Discussion
Based on our experiments, we have developed a model with promising performance. This model achieved over 90% accuracy on our test set and has demonstrated good results in practical applications.
ποΈ Training
Please ensure that you've built the foundational image docsaidtrainingbase_image from DocsaidKit.
ποΈ Submission
The real world is full of surprises, and you're bound to encounter situations where things don't quite fit.