锦中人工智能助手

pip install transformers torch

from transformers import AutoTokenizer, AutoModelForQuestionAnswering

import torch

tokenizer = AutoTokenizer.from_pretrained("deepset/roberta-base-squad2")

model = AutoModelForQuestionAnswering.from_pretrained("deepset/roberta-base-squad2")

question = "谁是美国第一位总统？"

context = "乔治·华盛顿是美国的第一位总统，他在1789年就职。"

inputs = tokenizer(question, context, return_tensors="pt")

outputs = model(**inputs)

answer_start_index = outputs.start_logits.argmax()

answer_end_index = outputs.end_logits.argmax()

predict_answer_tokens = inputs["input_ids"][0][answer_start_index:answer_end_index + 1]

answer = tokenizer.decode(predict_answer_tokens)

print("答案是：", answer)

import torch

import torch.nn as nn

import torch.optim as optim

# 简单的问答模型

class QuestionAnswerModel(nn.Module):

def __init__(self, vocab_size, embedding_dim, hidden_dim):

super(QuestionAnswerModel, self).__init__()

self.embedding = nn.Embedding(vocab_size, embedding_dim)

self.lstm = nn.LSTM(embedding_dim, hidden_dim, batch_first=True)

self.output_layer = nn.Linear(hidden_dim, 1)

def forward(self, x):

embedded = self.embedding(x)

lstm_out, _ = self.lstm(embedded)

out = self.output_layer(lstm_out)

return out

# 假设数据

vocab_size = 10000

embedding_dim = 128

hidden_dim = 256

model = QuestionAnswerModel(vocab_size, embedding_dim, hidden_dim)

criterion = nn.MSELoss()

optimizer = optim.Adam(model.parameters(), lr=0.001)

# 假设输入数据

input_data = torch.randint(0, vocab_size, (1, 10)) # 一批输入数据

labels = torch.tensor([[1.0]]) # 假设标签

# 训练循环

for epoch in range(10):

optimizer.zero_grad()

outputs = model(input_data)

loss = criterion(outputs, labels)

loss.backward()

optimizer.step()

print(f"Epoch {epoch+1}, Loss: {loss.item()}")