Colbert ModernBERT Base Turkish Uncased

🚀 土耳其长上下文基于ColBERT的重排器

这是一个基于 PyLate 库微调的模型，微调基础模型为 99eren99/ModernBERT-base-Turkish-uncased-mlm。该模型可将句子和段落映射为128维的密集向量序列，通过MaxSim算子进行语义文本相似度计算。

✨ 主要特性

• 基于 ColBERT 架构，适用于土耳其语长上下文重排任务。
• 可将文本映射为128维密集向量，用于语义相似度计算。

📦 安装指南

首先安装 PyLate 库：

pip install -U einops flash_attn
pip install -U pylate

然后对文本进行归一化处理：lambda x: x.replace("İ", "i").replace("I", "ı").lower()

💻 使用示例

基础用法

文档索引

from pylate import indexes, models, retrieve

# Step 1: Load the ColBERT model
document_length = 180#some integer [0,8192] for truncating documents, you can maybe try rope scaling for longer inputs  
model = models.ColBERT(
    model_name_or_path="99eren99/ColBERT-ModernBERT-base-Turkish-uncased", document_length=document_length
)
try:
    model.tokenizer.model_input_names.remove("token_type_ids")
except:
    pass
#model.to("cuda")

# Step 2: Initialize the Voyager index
index = indexes.Voyager(
    index_folder="pylate-index",
    index_name="index",
    override=True,  # This overwrites the existing index if any
)

# Step 3: Encode the documents
documents_ids = ["1", "2", "3"]
documents = ["document 1 text", "document 2 text", "document 3 text"]

documents_embeddings = model.encode(
    documents,
    batch_size=32,
    is_query=False,  # Ensure that it is set to False to indicate that these are documents, not queries
    show_progress_bar=True,
)

# Step 4: Add document embeddings to the index by providing embeddings and corresponding ids
index.add_documents(
    documents_ids=documents_ids,
    documents_embeddings=documents_embeddings,
)

注意，无需每次都重新创建索引和编码文档。创建索引并添加文档后，可通过以下方式加载并复用索引：

# To load an index, simply instantiate it with the correct folder/name and without overriding it
index = indexes.Voyager(
    index_folder="pylate-index",
    index_name="index",
)

查询文档

# Step 1: Initialize the ColBERT retriever
retriever = retrieve.ColBERT(index=index)

# Step 2: Encode the queries
queries_embeddings = model.encode(
    ["query for document 3", "query for document 1"],
    batch_size=32,
    is_query=True,  #  # Ensure that it is set to False to indicate that these are queries
    show_progress_bar=True,
)

# Step 3: Retrieve top-k documents
scores = retriever.retrieve(
    queries_embeddings=queries_embeddings, 
    k=10,  # Retrieve the top 10 matches for each query
)

重排文档

from pylate import rank, models

queries = [
    "query A",
    "query B",
]

documents = [
    ["document A", "document B"],
    ["document 1", "document C", "document B"],
]

documents_ids = [
    [1, 2],
    [1, 3, 2],
]

model = models.ColBERT(
    model_name_or_path=pylate_model_id,
)

queries_embeddings = model.encode(
    queries,
    is_query=True,
)

documents_embeddings = model.encode(
    documents,
    is_query=False,
)

reranked_documents = rank.rerank(
    documents_ids=documents_ids,
    queries_embeddings=queries_embeddings,
    documents_embeddings=documents_embeddings,
)

📚 详细文档

模型来源

• 文档说明：PyLate 文档
• 代码仓库：GitHub 上的 PyLate
• Hugging Face：Hugging Face 上的 PyLate 模型

评估结果

长上下文后期交互检索模型的 nDCG 和召回率得分，测试代码和详细指标见 "./assets"

📄 许可证

本项目采用 Apache-2.0 许可证。