README.txt

# Uncased Finnish Sentence BERT model

Finnish Sentence BERT trained from FinBERT

## Training

    FinBERT model: TurkuNLP/bert-base-finnish-uncased-v1
    Data: The data provided [here] (https://turkunlp.org/paraphrase.html), including the Finnish Paraphrase Corpus and the automatically collected paraphrase candidates (500K positive and 5M negative)
    Pooling: mean pooling
    Task: Binary prediction, whether two sentences are paraphrases or not. Note: the labels 3 and 4 are considered paraphrases, and labels 1 and 2 non-paraphrases. [Details on labels] (https://aclanthology.org/2021.nodalida-main.29/)

## Usage

The same as in [HuggingFace documentation] (https://huggingface.co/sentence-transformers/bert-base-nli-mean-tokens). Either through `SentenceTransformer` or `HuggingFace Transformers`

### SentenceTransformer

```
from sentence_transformers import SentenceTransformer
sentences = ["Tämä on esimerkkilause.", "Tämä on toinen lause."]

model = SentenceTransformer('TurkuNLP/sbert-uncased-finnish-paraphrase')
embeddings = model.encode(sentences)
print(embeddings)
```

### HuggingFace Transformers

```
from transformers import AutoTokenizer, AutoModel
import torch


#Mean Pooling - Take attention mask into account for correct averaging
def mean_pooling(model_output, attention_mask):
    token_embeddings = model_output[0] #First element of model_output contains all token embeddings
    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 we want sentence embeddings for
sentences = ["Tämä on esimerkkilause.", "Tämä on toinen lause."]

# Load model from HuggingFace Hub
tokenizer = AutoTokenizer.from_pretrained('TurkuNLP/sbert-uncased-finnish-paraphrase')
model = AutoModel.from_pretrained('TurkuNLP/sbert-uncased-finnish-paraphrase')

# Tokenize sentences
encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt')

# Compute token embeddings
with torch.no_grad():
    model_output = model(**encoded_input)

# Perform pooling. In this case, mean pooling.
sentence_embeddings = mean_pooling(model_output, encoded_input['attention_mask'])

print("Sentence embeddings:")
print(sentence_embeddings)
```