锦中融合门户系统

import torch

from transformers import AutoTokenizer, AutoModelForQuestionAnswering, Trainer, TrainingArguments

from datasets import load_dataset

# 加载预训练模型和分词器

model_name = "bert-base-uncased"

tokenizer = AutoTokenizer.from_pretrained(model_name)

model = AutoModelForQuestionAnswering.from_pretrained(model_name)

# 加载商标相关的数据集（这里假设有一个本地文件）

dataset = load_dataset("csv", data_files="trademark_data.csv")

# 对数据进行预处理

def preprocess_function(examples):

return tokenizer(

examples["question"],

examples["context"],

truncation=True,

padding="max_length",

max_length=512,

return_offsets_mapping=True

)

tokenized_datasets = dataset.map(preprocess_function, batched=True)

# 定义训练参数

training_args = TrainingArguments(

output_dir="./results",

evaluation_strategy="epoch",

learning_rate=2e-5,

per_device_train_batch_size=16,

num_train_epochs=3,

weight_decay=0.01,

)

# 定义Trainer

trainer = Trainer(

model=model,

args=training_args,

train_dataset=tokenized_datasets["train"],

eval_dataset=tokenized_datasets["validation"],

)

# 开始训练

trainer.train()

from sentence_transformers import SentenceTransformer, util

# 加载预训练的Sentence-BERT模型

model = SentenceTransformer('all-MiniLM-L6-v2')

# 两个商标名称

trademark1 = "Apple Inc."

trademark2 = "Applesauce Co."

# 生成嵌入向量

embeddings1 = model.encode(trademark1, convert_to_tensor=True)

embeddings2 = model.encode(trademark2, convert_to_tensor=True)

# 计算余弦相似度

similarity = util.cos_sim(embeddings1, embeddings2)

print(f"Similarity between '{trademark1}' and '{trademark2}': {similarity.item():.4f}")