jegorkitskerkin/robbert-v2-dutch-base-mqa-finetuned

这是一个 sentence-transformers 模型：它将句子和段落映射到一个768维的稠密向量空间，可用于聚类或语义搜索等任务。

这个模型是 pdelobelle/robbert-v2-dutch-base 的微调版本。它是在 clips/mqa 的100万行荷兰语常见问题解答对上进行微调的。

使用方法（Sentence-Transformers）

当你安装了 sentence-transformers 后，使用这个模型就变得很容易：

pip install -U sentence-transformers

然后你可以像这样使用模型：

from sentence_transformers import SentenceTransformer
sentences = ["This is an example sentence", "Each sentence is converted"]

model = SentenceTransformer('jegorkitskerkin/robbert-v2-dutch-base-mqa-finetuned')
embeddings = model.encode(sentences)
print(embeddings)

使用方法（HuggingFace Transformers）

如果没有 sentence-transformers ，你可以这样使用模型：首先，将输入通过变换器模型传递，然后必须在上下文词嵌入之上应用正确的池化操作。

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 = ['This is an example sentence', 'Each sentence is converted']

# Load model from HuggingFace Hub
tokenizer = AutoTokenizer.from_pretrained('jegorkitskerkin/robbert-v2-dutch-base-mqa-finetuned')
model = AutoModel.from_pretrained('jegorkitskerkin/robbert-v2-dutch-base-mqa-finetuned')

# 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)

训练

该模型使用以下参数进行训练：

DataLoader :

torch.utils.data.dataloader.DataLoader : 长度为12500，参数如下：

{'batch_size': 80, 'sampler': 'torch.utils.data.sampler.RandomSampler', 'batch_sampler': 'torch.utils.data.sampler.BatchSampler'}

损失函数 :

sentence_transformers.losses.MultipleNegativesRankingLoss.MultipleNegativesRankingLoss : 参数如下：

{'scale': 20.0, 'similarity_fct': 'cos_sim'}

fit()方法的参数：

{
    "epochs": 3,
    "evaluation_steps": 0,
    "evaluator": "NoneType",
    "max_grad_norm": 1,
    "optimizer_class": "<class 'transformers.optimization.AdamW'>",
    "optimizer_params": {
        "lr": 2e-05
    },
    "scheduler": "WarmupLinear",
    "steps_per_epoch": null,
    "warmup_steps": 10000,
    "weight_decay": 0.01
}

完整模型架构

SentenceTransformer(
  (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: RobertaModel 
  (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False})
)

引用和作者