|
--- |
|
license: cc-by-4.0 |
|
--- |
|
|
|
# TACDEC-model |
|
This is a simple model with weights and reproducible code for the results in TACDEC-paper. |
|
|
|
What you can find in this repo is: |
|
- The simple [model](https://huggingface.co./SimulaMet-HOST/TACDEC-model/resolve/main/model.py?download=true) used in the TACDEC-paper |
|
- The [weights](https://huggingface.co./SimulaMet-HOST/TACDEC-model/resolve/main/simple_model_weights.pt?download=true) used in the proof-of-concept section in the TACDEC-paper |
|
- A first notebook, [feature_extraction.ipynb](https://huggingface.co./SimulaMet-HOST/TACDEC-model/resolve/main/feature_extraction.ipynb?download=true), that contains a feature extraction process using DINOv2. |
|
- A second notebook, [train_classifier.ipynb](https://huggingface.co./SimulaMet-HOST/TACDEC-model/resolve/main/train_classifier.ipynb?download=true), that uses the features that were either extracted using the first notebook, or downloaded directly from [TACDEC repo](https://huggingface.co./datasets/SimulaMet-HOST/TACDEC). |
|
|
|
We highly recommend downloading the already extracted and concatenated (features)[https://huggingface.co./datasets/SimulaMet-HOST/TACDEC/resolve/main/sorted_cls_tokens_features.pt] and the concatenated (labels)[https://huggingface.co./datasets/SimulaMet-HOST/TACDEC/resolve/main/sorted_cls_tokens_labels.npy] if you wish to try the dataset/model. You would then just have to run the second notebook. |
|
|
|
|
|
If you hold more interest in DINOv2, the **feature_extraction.ipynb** could hold good value. |
|
|
|
In both notebooks, there should be good enough documentation, but should you have any questions, see [TACDEC](https://huggingface.co./datasets/SimulaMet-HOST/TACDEC). |
|
|
|
## More information |
|
For any other information or information about the dataset: |
|
[TACDEC](https://huggingface.co./datasets/SimulaMet-HOST/TACDEC) |
|
|
