Translation
Transformers
Safetensors
PyTorch
Russian
Kazakh
t5
text2text-generation
russian
kazakh
text-generation-inference
Instructions to use deepvk/kazRush-kk-ru with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use deepvk/kazRush-kk-ru 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="deepvk/kazRush-kk-ru")# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForSeq2SeqLM tokenizer = AutoTokenizer.from_pretrained("deepvk/kazRush-kk-ru") model = AutoModelForSeq2SeqLM.from_pretrained("deepvk/kazRush-kk-ru", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Commit ·
c07b0f7
1
Parent(s): f25ad80
readme update #2
Browse files
README.md
CHANGED
|
@@ -16,7 +16,7 @@ language:
|
|
| 16 |
|
| 17 |
# kazRush-kk-ru
|
| 18 |
|
| 19 |
-
KazRush is a translation model for translating from Kazakh to Russian.
|
| 20 |
|
| 21 |
## Usage
|
| 22 |
|
|
@@ -36,7 +36,7 @@ model = AutoModelForSeq2SeqLM.from_pretrained('deepvk/kazRush-kk-ru')
|
|
| 36 |
tokenizer = AutoTokenizer.from_pretrained('deepvk/kazRush-kk-ru')
|
| 37 |
|
| 38 |
def generate(text, **kwargs):
|
| 39 |
-
inputs = tokenizer(text, return_tensors='pt')
|
| 40 |
with torch.no_grad():
|
| 41 |
hypotheses = model.generate(**inputs, num_beams=5, **kwargs)
|
| 42 |
return tokenizer.decode(hypotheses[0], skip_special_tokens=True)
|
|
@@ -72,7 +72,7 @@ Calculation of BLEU and chrF follows the standart implementation from [sacreBLEU
|
|
| 72 |
| [nllb-200-3.3B](https://huggingface.co/facebook/nllb-200-3.3B) | 3.3B | 21.5 | 50.7 | 0.8874 |
|
| 73 |
| [our model (azi7oreu)]() | 64.8 M | 16.3 | 46.6 | 0.8428 |
|
| 74 |
| [our model (fnkx3n1x)]() | 64.8 M | 17.5 | 47.4 | 0.8029 |
|
| 75 |
-
| [our model (243xhibn)]() | 64.8 M | 17.
|
| 76 |
|
| 77 |
## Examples of usage (ПЕРЕДЕЛАТЬ НА ЛУЧШУЮ, пока что тут примеры с azi7oreu):
|
| 78 |
|
|
|
|
| 16 |
|
| 17 |
# kazRush-kk-ru
|
| 18 |
|
| 19 |
+
KazRush-kk-ru is a translation model for translating from Kazakh to Russian.
|
| 20 |
|
| 21 |
## Usage
|
| 22 |
|
|
|
|
| 36 |
tokenizer = AutoTokenizer.from_pretrained('deepvk/kazRush-kk-ru')
|
| 37 |
|
| 38 |
def generate(text, **kwargs):
|
| 39 |
+
inputs = tokenizer(text, return_tensors='pt').to('cuda')
|
| 40 |
with torch.no_grad():
|
| 41 |
hypotheses = model.generate(**inputs, num_beams=5, **kwargs)
|
| 42 |
return tokenizer.decode(hypotheses[0], skip_special_tokens=True)
|
|
|
|
| 72 |
| [nllb-200-3.3B](https://huggingface.co/facebook/nllb-200-3.3B) | 3.3B | 21.5 | 50.7 | 0.8874 |
|
| 73 |
| [our model (azi7oreu)]() | 64.8 M | 16.3 | 46.6 | 0.8428 |
|
| 74 |
| [our model (fnkx3n1x)]() | 64.8 M | 17.5 | 47.4 | 0.8029 |
|
| 75 |
+
| [our model (243xhibn)]() | 64.8 M | 17.4 | 47.4 | 0.8556 |
|
| 76 |
|
| 77 |
## Examples of usage (ПЕРЕДЕЛАТЬ НА ЛУЧШУЮ, пока что тут примеры с azi7oreu):
|
| 78 |
|