Slavic Byt5 for Kashubian Inflection
Collection
Please refer to https://github.com/TUM-NLP/kashubian-morphological-inflection for a model description. โข 14 items โข Updated
How to use livles/hier-csb-pol with Transformers:
# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
tokenizer = AutoTokenizer.from_pretrained("livles/hier-csb-pol")
model = AutoModelForSeq2SeqLM.from_pretrained("livles/hier-csb-pol", device_map="auto")This model is a fine-tuned version of google/byt5-small on an unknown dataset. It achieves the following results on the evaluation set:
More information needed
More information needed
More information needed
The following hyperparameters were used during training:
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.5387 | 1.0 | 1250 | 0.3407 |
| 0.2951 | 2.0 | 2500 | 0.2344 |
| 0.2427 | 3.0 | 3750 | 0.2155 |
| 0.2058 | 4.0 | 5000 | 0.1983 |
| 0.1898 | 5.0 | 6250 | 0.2020 |
| 0.1710 | 6.0 | 7500 | 0.1922 |
| 0.1618 | 7.0 | 8750 | 0.1962 |
| 0.1574 | 8.0 | 10000 | 0.1946 |
| 0.1530 | 9.0 | 11250 | 0.1960 |
| 0.1399 | 10.0 | 12500 | 0.1982 |
| 0.1438 | 11.0 | 13750 | 0.1946 |
| 0.1370 | 12.0 | 15000 | 0.1966 |
Base model
google/byt5-small