365天深度学习打卡第N2周：中文文本分类-Pytorch实现

这里写自定义目录标题目标Before do anything :¶mount google drive by press mount button since my account has advanced protection¶导入数据构造数据集迭代器¶数据预处理¶生成数据批次和迭代器¶定义模型¶定义实例定义训练函数与评估函数¶拆分数据集并运行模型¶使用测试数据集评估模型¶🍨 本文为🔗365

WTIAW.TIAW

528人浏览 · 2023-05-19 15:12:49

WTIAW.TIAW · 2023-05-19 15:12:49 发布

这里写自定义目录标题

目标
Before do anything :¶
- mount google drive by press mount button since my account has advanced protection¶
导入数据
构造数据集迭代器¶
数据预处理¶
生成数据批次和迭代器¶
定义模型¶
定义实例
定义训练函数与评估函数¶
拆分数据集并运行模型¶
使用测试数据集评估模型¶

🍨 本文为🔗365天深度学习训练营中的学习记录博客

🍖 原作者：K同学啊|接辅导、项目定制

目标

中文文本的处理

Before do anything :¶

mount google drive by press mount button since my account has advanced protection¶

install dependency
save env requirements

!rsync -avhPt /content/drive/MyDrive/Colab\ Notebooks/NoteBooks/study/365_days_for_deeplearn_NLP/* .
!pip install torchviz
!pip install torch torchvision torchaudio torchinfo
!pip install pytorch-lightning
!pip install torchtext portalocker
!pip freeze > colab_N1_requirements_`date +%Y%m%d`.txt
!cp colab_N1_requirements_`date +%Y%m%d`.txt /content/drive/MyDrive/Colab\ Notebooks/NoteBooks/study/365_days_for_deeplearn_NLP/ -v

workspace_dir="/content/drive/MyDrive/Colab Notebooks/NoteBooks/study/365_days_for_deeplearn_NLP/"

import torch
print(torch.__version__)

2.0.0+cu118

import torch
import torch.nn as nn
import torchvision.transforms as transforms
import torchvision
from torchvision import transforms, datasets
import torch.nn.functional as F

device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
device

device(type=‘cuda’)

导入数据

import pandas as pd
train_data = pd.read_csv('./train_n2.csv',sep='\t',header=None)
train_data.head()

0 1
0 还有双鸭山到淮阴的汽车票吗13号的 Travel-Query
1 从这里怎么回家 Travel-Query
2 随便播放一首专辑阁楼里的佛里的歌 Music-Play
3 给看一下墓王之王嘛 FilmTele-Play
4 我想看挑战两把s686打突变团竞的游戏视频 Video-Play

构造数据集迭代器¶

def coustom_data_iter(text, labels):
    for x,y in zip(text,labels):
        yield x,y
train_iter = coustom_data_iter(train_data[0].values[:],train_data[1].values[:])

数据预处理¶

from torchtext.data.utils import get_tokenizer
from torchtext.vocab import build_vocab_from_iterator
import jieba
tokenizer = jieba.lcut

def yield_tokens(data_iter):
    for text,_ in data_iter:
        yield tokenizer(text)
vocab = build_vocab_from_iterator(yield_tokens(train_iter), specials=["<unk>"])
vocab.set_default_index(vocab["<unk>"])

Building prefix dict from the default dictionary …
DEBUG:jieba:Building prefix dict from the default dictionary …
Dumping model to file cache /tmp/jieba.cache
DEBUG:jieba:Dumping model to file cache /tmp/jieba.cache
Loading model cost 0.710 seconds.
DEBUG:jieba:Loading model cost 0.710 seconds.
Prefix dict has been built successfully.
DEBUG:jieba:Prefix dict has been built successfully.

vocab(['我','想','看','和平','精英','上','战神','必备','技巧','的','游戏','视频'])

[2, 10, 13, 973, 1079, 146, 7724, 7574, 7793, 1, 186, 28]

label_name = list(set(train_data[1].values[:]))
print(label_name)

[‘FilmTele-Play’, ‘Video-Play’, ‘Calendar-Query’, ‘Alarm-Update’, ‘HomeAppliance-Control’, ‘Music-Play’, ‘Travel-Query’, ‘Weather-Query’, ‘Audio-Play’, ‘Other’, ‘TVProgram-Play’, ‘Radio-Listen’]

text_pipeline = lambda x: vocab(tokenizer(x))
label_pipeline = lambda x: label_name.index(x)

print(text_pipeline('我想看和平精英上战神必备技巧的游戏视频'))
print(label_pipeline('Video-Play'))

[2, 10, 13, 973, 1079, 146, 7724, 7574, 7793, 1, 186, 28]
1

生成数据批次和迭代器¶

这个是上次学习的代码

from torch.utils.data import DataLoader

def collate_batch(batch):
    label_list, text_list, offsets = [], [], [0]
    
    for (_text,_label) in batch:
        # 标签列表
        label_list.append(label_pipeline(_label))
        
        # 文本列表
        processed_text = torch.tensor(text_pipeline(_text), dtype=torch.int64)
        text_list.append(processed_text)
        
        # 偏移量，即语句的总词汇量
        offsets.append(processed_text.size(0))
        
    label_list = torch.tensor(label_list, dtype=torch.int64)
    text_list  = torch.cat(text_list)
    offsets    = torch.tensor(offsets[:-1]).cumsum(dim=0) #返回维度dim中输入元素的累计和
    
    return text_list.to(device),label_list.to(device), offsets.to(device)

# 数据加载器，调用示例
dataloader = DataLoader(train_iter,
                        batch_size=8,
                        shuffle   =False,
                        collate_fn=collate_batch)

定义模型¶

from torch import nn

class TextClassificationModel(nn.Module):

    def __init__(self, vocab_size, embed_dim, num_class):
        super(TextClassificationModel, self).__init__()
        
        self.embedding = nn.EmbeddingBag(vocab_size,   # 词典大小
                                         embed_dim,    # 嵌入的维度
                                         sparse=False) # 
        
        self.fc = nn.Linear(embed_dim, num_class)
        self.init_weights()

    def init_weights(self):
        initrange = 0.5
        self.embedding.weight.data.uniform_(-initrange, initrange) # 初始化权重
        self.fc.weight.data.uniform_(-initrange, initrange)        
        self.fc.bias.data.zero_()                                  # 偏置值归零

    def forward(self, text, offsets):
        embedded = self.embedding(text, offsets)
        return self.fc(embedded)

定义实例

num_class  = len(label_name)
vocab_size = len(vocab)
em_size    = 64
model      = TextClassificationModel(vocab_size, em_size, num_class).to(device)

定义训练函数与评估函数¶

import time

def train(dataloader):
    model.train()  # 切换为训练模式
    total_acc, train_loss, total_count = 0, 0, 0
    log_interval = 50
    start_time   = time.time()

    for idx, (text,label,offsets) in enumerate(dataloader):
        
        predicted_label = model(text, offsets)
        
        optimizer.zero_grad()                    # grad属性归零
        loss = criterion(predicted_label, label) # 计算网络输出和真实值之间的差距，label为真实值
        loss.backward()                          # 反向传播
        torch.nn.utils.clip_grad_norm_(model.parameters(), 0.1) # 梯度裁剪
        optimizer.step()  # 每一步自动更新
        
        # 记录acc与loss
        total_acc   += (predicted_label.argmax(1) == label).sum().item()
        train_loss  += loss.item()
        total_count += label.size(0)
        
        if idx % log_interval == 0 and idx > 0:
            elapsed = time.time() - start_time
            print('| epoch {:1d} | {:4d}/{:4d} batches '
                  '| train_acc {:4.3f} train_loss {:4.5f}'.format(epoch, idx, len(dataloader),
                                              total_acc/total_count, train_loss/total_count))
            total_acc, train_loss, total_count = 0, 0, 0
            start_time = time.time()

def evaluate(dataloader):
    model.eval()  # 切换为测试模式
    total_acc, train_loss, total_count = 0, 0, 0

    with torch.no_grad():
        for idx, (text,label,offsets) in enumerate(dataloader):
            predicted_label = model(text, offsets)
            
            loss = criterion(predicted_label, label)  # 计算loss值
            # 记录测试数据
            total_acc   += (predicted_label.argmax(1) == label).sum().item()
            train_loss  += loss.item()
            total_count += label.size(0)
            
    return total_acc/total_count, train_loss/total_count

拆分数据集并运行模型¶

| {:4d}/{:4d} batches '
                  '| train_acc {:4.3f} train_loss {:4.5f}'.format(epoch, idx, len(dataloader),
                                              total_acc/total_count, train_loss/total_count))
            total_acc, train_loss, total_count = 0, 0, 0
            start_time = time.time()

def evaluate(dataloader):
    model.eval()  # 切换为测试模式
    total_acc, train_loss, total_count = 0, 0, 0

    with torch.no_grad():
        for idx, (text,label,offsets) in enumerate(dataloader):
            predicted_label = model(text, offsets)
            
            loss = criterion(predicted_label, label)  # 计算loss值
            # 记录测试数据
            total_acc   += (predicted_label.argmax(1) == label).sum().item()
            train_loss  += loss.item()
            total_count += label.size(0)
            
    return total_acc/total_count, train_loss/total_count
拆分数据集并运行模型
from torch.utils.data.dataset import random_split
from torchtext.data.functional import to_map_style_dataset
# 超参数
EPOCHS     = 100 # epoch
LR         = 5  # 学习率
BATCH_SIZE = 1024 # batch size for training

criterion = torch.nn.CrossEntropyLoss()
optimizer = torch.optim.SGD(model.parameters(), lr=LR)
scheduler = torch.optim.lr_scheduler.StepLR(optimizer, 1.0, gamma=0.1)
total_accu = None

# 构建数据集
train_iter = coustom_data_iter(train_data[0].values[:], train_data[1].values[:])
train_dataset = to_map_style_dataset(train_iter)

split_train_, split_valid_ = random_split(train_dataset,
                                          [int(len(train_dataset)*0.8),int(len(train_dataset)*0.2)])

train_dataloader = DataLoader(split_train_, batch_size=BATCH_SIZE,
                              shuffle=True, collate_fn=collate_batch)

valid_dataloader = DataLoader(split_valid_, batch_size=BATCH_SIZE,
                              shuffle=True, collate_fn=collate_batch)

for epoch in range(1, EPOCHS + 1):
    epoch_start_time = time.time()
    train(train_dataloader)
    val_acc, val_loss = evaluate(valid_dataloader)
    
    # 获取当前的学习率
    lr = optimizer.state_dict()['param_groups'][0]['lr']
    
    if total_accu is not None and total_accu > val_acc:
        scheduler.step()
    else:
        total_accu = val_acc
    print('-' * 69)
    print('| epoch {:1d} | time: {:4.2f}s | '
          'valid_acc {:4.3f} valid_loss {:4.3f} | lr {:4.6f}'.format(epoch,
                                           time.time() - epoch_start_time,
                                           val_acc,val_loss,lr))

    print('-' * 69)

---------------------------------------------------------------------
| epoch 1 | time: 1.83s | valid_acc 0.944 valid_loss 0.000 | lr 5.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 2 | time: 1.85s | valid_acc 0.942 valid_loss 0.000 | lr 5.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 3 | time: 1.40s | valid_acc 0.943 valid_loss 0.000 | lr 0.500000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 4 | time: 1.82s | valid_acc 0.943 valid_loss 0.000 | lr 0.050000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 5 | time: 1.95s | valid_acc 0.943 valid_loss 0.000 | lr 0.005000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 6 | time: 1.30s | valid_acc 0.943 valid_loss 0.000 | lr 0.000500
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 7 | time: 1.27s | valid_acc 0.943 valid_loss 0.000 | lr 0.000050
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 8 | time: 1.33s | valid_acc 0.943 valid_loss 0.000 | lr 0.000005
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 9 | time: 1.54s | valid_acc 0.943 valid_loss 0.000 | lr 0.000001
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 10 | time: 1.91s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 11 | time: 1.85s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 12 | time: 1.30s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 13 | time: 1.27s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 14 | time: 1.31s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 15 | time: 1.27s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 16 | time: 1.31s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 17 | time: 1.22s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 18 | time: 1.31s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 19 | time: 1.45s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 20 | time: 1.88s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 21 | time: 1.88s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 22 | time: 1.31s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 23 | time: 1.27s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 24 | time: 1.32s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 25 | time: 1.28s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 26 | time: 1.23s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 27 | time: 1.32s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 28 | time: 1.35s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 29 | time: 1.44s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 30 | time: 1.90s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 31 | time: 1.94s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 32 | time: 1.31s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 33 | time: 1.29s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 34 | time: 1.25s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 35 | time: 1.33s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 36 | time: 1.31s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 37 | time: 1.21s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 38 | time: 1.26s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 39 | time: 1.41s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 40 | time: 1.81s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 41 | time: 1.93s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 42 | time: 1.36s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 43 | time: 1.23s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 44 | time: 1.36s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 45 | time: 1.29s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 46 | time: 1.29s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 47 | time: 1.29s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 48 | time: 1.38s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 49 | time: 1.42s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 50 | time: 1.83s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 51 | time: 1.86s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 52 | time: 1.32s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 53 | time: 1.32s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 54 | time: 1.32s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 55 | time: 1.25s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 56 | time: 1.28s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 57 | time: 1.32s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 58 | time: 1.25s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 59 | time: 1.46s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 60 | time: 1.92s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 61 | time: 1.93s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 62 | time: 1.39s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 63 | time: 1.34s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 64 | time: 1.23s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 65 | time: 1.25s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 66 | time: 1.24s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 67 | time: 1.29s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 68 | time: 1.33s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 69 | time: 1.47s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 70 | time: 1.90s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 71 | time: 1.88s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 72 | time: 1.26s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 73 | time: 1.25s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 74 | time: 1.33s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 75 | time: 1.28s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 76 | time: 1.30s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 77 | time: 1.30s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 78 | time: 1.28s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 79 | time: 1.36s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 80 | time: 1.90s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 81 | time: 1.91s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 82 | time: 1.30s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 83 | time: 1.28s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 84 | time: 1.30s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 85 | time: 1.27s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 86 | time: 1.29s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 87 | time: 1.40s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 88 | time: 1.26s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 89 | time: 1.38s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 90 | time: 1.87s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 91 | time: 1.87s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 92 | time: 1.31s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 93 | time: 1.24s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 94 | time: 1.26s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 95 | time: 1.32s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 96 | time: 1.30s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 97 | time: 1.23s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 98 | time: 1.25s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 99 | time: 1.21s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------
---------------------------------------------------------------------
| epoch 100 | time: 1.80s | valid_acc 0.943 valid_loss 0.000 | lr 0.000000
---------------------------------------------------------------------

使用测试数据集评估模型¶

def predict(text, text_pipeline):
    with torch.no_grad():
        text = torch.tensor(text_pipeline(text))
        output = model(text, torch.tensor([0]))
        return output.argmax(1).item()

ex_text_str = "怎么到吕凡超的家"

model = model.to("cpu")
print("该文本的类别是：%s" %label_name[predict(ex_text_str, text_pipeline)])

该文本的类别是：Travel-Query

技术共进，成长同行——讯飞AI开发者社区

更多推荐

什么是人工智能大模型？

讯飞AI开发者社区

号课堂§2.2：声明范式

声明式语言——尤其是函数式语言和逻辑式语言——擅长基于数理逻辑的应用，如人工智能、符号处理、数据库、编译器等，对基于业务逻辑的、尤其是交互式或事件驱动型的应用就不那么得心应手了。命令式编程模拟电脑运算，是行动导向的，关键在于定义解法，即“怎么做”，因而算法是显性而目标是隐性的；声明式编程模拟人脑思维，是目标驱动的，关键在于描述问题，即“做什么”，因而目标是显性而算法是隐性的。”冒号娓娓道来，“除了