Instructions to use GottBERT/GottBERT_base_last with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use GottBERT/GottBERT_base_last with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("fill-mask", model="GottBERT/GottBERT_base_last")# Load model directly from transformers import AutoTokenizer, AutoModelForMaskedLM tokenizer = AutoTokenizer.from_pretrained("GottBERT/GottBERT_base_last") model = AutoModelForMaskedLM.from_pretrained("GottBERT/GottBERT_base_last") - Inference
- Notebooks
- Google Colab
- Kaggle
Raphael Scheible commited on
Update README.md
Browse files
README.md
CHANGED
|
@@ -94,6 +94,9 @@ Details:
|
|
| 94 |
- **Filtered vs Unfiltered Data**: Minor improvements seen with filtered data, but not significant enough to justify filtering in every case.
|
| 95 |
- **Computation Limitations**: Fixed memory allocation on TPUs required processing data as a single stream, unlike GPU training which preserves document boundaries. Training was performed in 32-bit mode due to framework limitations, increasing memory usage.
|
| 96 |
|
|
|
|
|
|
|
|
|
|
| 97 |
## Citations
|
| 98 |
If you use GottBERT in your research, please cite the following paper:
|
| 99 |
```bibtex
|
|
|
|
| 94 |
- **Filtered vs Unfiltered Data**: Minor improvements seen with filtered data, but not significant enough to justify filtering in every case.
|
| 95 |
- **Computation Limitations**: Fixed memory allocation on TPUs required processing data as a single stream, unlike GPU training which preserves document boundaries. Training was performed in 32-bit mode due to framework limitations, increasing memory usage.
|
| 96 |
|
| 97 |
+
## Fairseq Checkpoints
|
| 98 |
+
Get the fairseq checkpoints [here](https://drive.proton.me/urls/CFSGE8ZK9R#1F1G727lv77k).
|
| 99 |
+
|
| 100 |
## Citations
|
| 101 |
If you use GottBERT in your research, please cite the following paper:
|
| 102 |
```bibtex
|