🚀 Turkish Mini BERT Uncased Mean NLI STSB TR
This model maps sentences and paragraphs to a 256-dimensional dense vector space, suitable for tasks like clustering and semantic search.
This is a sentence-transformers model. It can transform sentences and paragraphs into a 256-dimensional dense vector space, which can be used for tasks such as clustering or semantic search.
This model is adapted from ytu-ce-cosmos/turkish-mini-bert-uncased and fine - tuned on the following datasets:
⚠️ Important Note
As stated by the model's authors, all texts need to be manually lowercased:
text.replace("I", "ı").lower()
🚀 Quick Start
✨ Features
- Maps sentences and paragraphs to a 256 - dimensional dense vector space.
- Can be used for clustering and semantic search tasks.
📦 Installation
Using this model becomes easy when you have sentence-transformers installed:
pip install -U sentence-transformers
💻 Usage Examples
Basic Usage (Sentence - Transformers)
from sentence_transformers import SentenceTransformer
sentences = ["Bu örnek bir cümle", "Her cümle dönüştürülür"]
model = SentenceTransformer('atasoglu/turkish-mini-bert-uncased-mean-nli-stsb-tr')
embeddings = model.encode(sentences)
print(embeddings)
Advanced Usage (HuggingFace Transformers)
Without sentence-transformers, you can use the model like this: First, you pass your input through the transformer model, then you have to apply the right pooling - operation on - top of the contextualized word embeddings.
from transformers import AutoTokenizer, AutoModel
import torch
def mean_pooling(model_output, attention_mask):
token_embeddings = model_output[0]
input_mask_expanded = attention_mask.unsqueeze(-1).expand(token_embeddings.size()).float()
return torch.sum(token_embeddings * input_mask_expanded, 1) / torch.clamp(input_mask_expanded.sum(1), min=1e-9)
sentences = ["Bu örnek bir cümle", "Her cümle dönüştürülür"]
tokenizer = AutoTokenizer.from_pretrained('atasoglu/turkish-mini-bert-uncased-mean-nli-stsb-tr')
model = AutoModel.from_pretrained('atasoglu/turkish-mini-bert-uncased-mean-nli-stsb-tr')
encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt')
with torch.no_grad():
model_output = model(**encoded_input)
sentence_embeddings = mean_pooling(model_output, encoded_input['attention_mask'])
print("Sentence embeddings:")
print(sentence_embeddings)
📚 Documentation
Evaluation Results
Achieved results on the STS - b test split are given below:
Cosine - Similarity : Pearson: 0.8117 Spearman: 0.8074
Manhattan - Distance: Pearson: 0.8029 Spearman: 0.7972
Euclidean - Distance: Pearson: 0.8028 Spearman: 0.7977
Dot - Product - Similarity: Pearson: 0.7563 Spearman: 0.7435
Training
The model was trained with the parameters:
DataLoader:
torch.utils.data.dataloader.DataLoader
of length 45 with parameters:
{'batch_size': 128, 'sampler': 'torch.utils.data.sampler.RandomSampler', 'batch_sampler': 'torch.utils.data.sampler.BatchSampler'}
Loss:
sentence_transformers.losses.CosineSimilarityLoss.CosineSimilarityLoss
Parameters of the fit() - Method:
{
"epochs": 10,
"evaluation_steps": 4,
"evaluator": "sentence_transformers.evaluation.EmbeddingSimilarityEvaluator.EmbeddingSimilarityEvaluator",
"max_grad_norm": 1,
"optimizer_class": "<class 'torch.optim.adamw.AdamW'>",
"optimizer_params": {
"lr": 2e-05
},
"scheduler": "WarmupLinear",
"steps_per_epoch": null,
"warmup_steps": 45,
"weight_decay": 0.01
}
Full Model Architecture
SentenceTransformer(
(0): Transformer({'max_seq_length': 256, 'do_lower_case': False}) with Transformer model: BertModel
(1): Pooling({'word_embedding_dimension': 256, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False})
)
📄 License
This model is licensed under the MIT license.
📋 Information Table
Property |
Details |
Pipeline Tag |
Sentence - Similarity |
Tags |
sentence - transformers, feature - extraction, sentence - similarity, transformers |
Datasets |
nli_tr, emrecan/stsb - mt - turkish |
Library Name |
sentence - transformers |
Base Model |
ytu - ce - cosmos/turkish - mini - bert - uncased |
License |
MIT |