Instructions to use buddhist-nlp/mt5-3B-tib2eng with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use buddhist-nlp/mt5-3B-tib2eng with Transformers:
# Load model directly from transformers import AutoTokenizer, AutoModelForMultimodalLM tokenizer = AutoTokenizer.from_pretrained("buddhist-nlp/mt5-3B-tib2eng") model = AutoModelForMultimodalLM.from_pretrained("buddhist-nlp/mt5-3B-tib2eng") - Notebooks
- Google Colab
- Kaggle
Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
datasets:
|
| 3 |
+
- buddhist-nlp/tib_eng_bitext
|
| 4 |
+
language:
|
| 5 |
+
- bo
|
| 6 |
+
- en
|
| 7 |
+
metrics:
|
| 8 |
+
- bleu
|
| 9 |
+
- chrf
|
| 10 |
+
library_name: transformers
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
|Training setup | |
|
| 14 |
+
|----------------------|-------|
|
| 15 |
+
|Num train steps | 10000|
|
| 16 |
+
|Max seq len | 256|
|
| 17 |
+
|Batch size | 512|
|
| 18 |
+
|Total data points seen|5.1 mil|
|
| 19 |
+
|Total tokens seen |450 mil|
|
| 20 |
+
|Checkpoint step | 8800|
|
| 21 |
+
|Learning rate | 3e-4|
|
| 22 |
+
|
| 23 |
+
|
| 24 |
+
|Metric|Val |Test|
|
| 25 |
+
|------|----|----|
|
| 26 |
+
|BLEU |30.0|27.3|
|
| 27 |
+
|chrf++|48.2|46.3|
|