Downloads · 30 days
99
8% of all-time downloads
patent-cpc-predictor
bradgrimm/patent-cpc-predictor
patent-cpc-predictor is a feature extraction model from bradgrimm. Use it when you need embeddings to search or compare text. It is set up for transformers. The card lists the license as mit.
This is a fine-tuned version of microsoft/deberta-v3-small for predicting Patent CPC codes.
All-time downloads
1.3K
Public
Parameters
141M
1.7 GB on disk
Likes
1
Public
Repo makeup
Click a slice to open those files.
.bin565 MB · 50%
Parameter types
How the weights are stored.
F32141M · 100%
Model card
From the Hugging Face model README
Patent CPC Predictor
This is a fine-tuned version of microsoft/deberta-v3-small for predicting Patent CPC codes.
Dataset
Dataset consists of titles and abstracts sampled from granted patent applications: https://www.kaggle.com/datasets/grimmace/sampled-patent-titles
Results
| Category | Accuracy |
|---|---|
| Section | 92% |
| Class | 88% |
| Subclass | 85% |
- Task
- Feature Extraction
- Library
- transformers
- Type
- deberta-v2
- License
- mit
- Languages
- en
- Created
- Jun 2, 2022
- Updated
- Jan 31, 2024