Translation
Transformers
PyTorch
English
Hindi
viuai
viutranslate
sarus-500m
nmt
english-to-hindi
hindi-to-english
indic
devanagari
bfloat16
zero-synthetic
Instructions to use ViuAI/ViuTranslate with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use ViuAI/ViuTranslate with Transformers:
# Use a pipeline as a high-level helper # Warning: Pipeline type "translation" is no longer supported in transformers v5. # You must load the model directly (see below) or downgrade to v4.x with: # pip install "transformers<5.0.0" from transformers import pipeline pipe = pipeline("translation", model="ViuAI/ViuTranslate")# pip install -U transformers accelerate # Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("ViuAI/ViuTranslate", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Upload train_viutranslate.py with huggingface_hub
Browse files- train_viutranslate.py +2 -2
train_viutranslate.py
CHANGED
|
@@ -62,8 +62,8 @@ def auto_profile_hardware():
|
|
| 62 |
dtype = torch.bfloat16 if bf16_supported else torch.float16
|
| 63 |
|
| 64 |
if vram_gb >= 30:
|
| 65 |
-
micro_batch, grad_accum =
|
| 66 |
-
desc = "
|
| 67 |
elif vram_gb >= 20:
|
| 68 |
micro_batch, grad_accum = 12, 6
|
| 69 |
desc = "Pro-Tier GPU (24GB)"
|
|
|
|
| 62 |
dtype = torch.bfloat16 if bf16_supported else torch.float16
|
| 63 |
|
| 64 |
if vram_gb >= 30:
|
| 65 |
+
micro_batch, grad_accum = 32, 2
|
| 66 |
+
desc = "NVIDIA RTX 5090 / High-Tier 32GB Blackwell Beast"
|
| 67 |
elif vram_gb >= 20:
|
| 68 |
micro_batch, grad_accum = 12, 6
|
| 69 |
desc = "Pro-Tier GPU (24GB)"
|