Feature Extraction
sentence-transformers
Safetensors
Transformers
Russian
English
t5
mteb
Eval Results (legacy)
Instructions to use ai-forever/FRIDA with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- sentence-transformers
How to use ai-forever/FRIDA with sentence-transformers:
from sentence_transformers import SentenceTransformer model = SentenceTransformer("ai-forever/FRIDA") sentences = [ "The weather is lovely today.", "It's so sunny outside!", "He drove to the stadium." ] embeddings = model.encode(sentences) similarities = model.similarity(embeddings, embeddings) print(similarities.shape) # [3, 3] - Transformers
How to use ai-forever/FRIDA with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("feature-extraction", model="ai-forever/FRIDA")# Load model directly from transformers import AutoTokenizer, AutoModel tokenizer = AutoTokenizer.from_pretrained("ai-forever/FRIDA") model = AutoModel.from_pretrained("ai-forever/FRIDA") - Inference
- Notebooks
- Google Colab
- Kaggle
model_FRIDA_static_avx512vnni_8bit.onnx
#7
by Colegero - opened
Квантизация выполнена с использованием 8-битной асимметричной квантизации весов (симметричной для активаций) и калибровочного датасета на основе подмножества датасета ai-forever/solyanka (конфигурация lenta, 100 сэмплов).
Тип квантизации: Статическая, 8-битная (AVX512 VNNI)
Калибровочный датасет: ai-forever/solyanka (lenta, 100 сэмплов)
Операторы для квантизации: Add, Gather, MatMul, Reshape, Softmax, Tanh
Метод калибровки: Entropy (256 бинов)
Используемые конфиги:
operators_to_quantize = ["Add", "Gather", "MatMul", "Reshape", "Softmax", "Tanh"]
config = AutoQuantizationConfig.avx512_vnni(
is_static=True,
use_symmetric_activations=False,
use_symmetric_weights=True,
per_channel=True,
operators_to_quantize=operators_to_quantize
)
return config
calibration_config = CalibrationConfig(
dataset_name="ai-forever/solyanka",
dataset_config_name="lenta",
dataset_split="train",
dataset_num_samples=num_calibration_samples,
method=CalibrationMethod.Entropy,
num_bins=256
)
Colegero changed pull request title from Upload model_FRIDA_static_avx512vnni_8bit.onnx to model_FRIDA_static_avx512vnni_8bit.onnx