Kaggle免费GPU使用攻略

公众号后台回复关键词： gpu ，获取B站视频演示教程。

一，注册Kaggle

在国内使用邮箱注册kaggle时会遇到一个人机验证的步骤，可以通过翻墙访问外网的方式完成，但比较麻烦。

推荐使用FireFox浏览器，下载Header Editor进行解决，无需翻墙相对简单。

1，下载安装FireFox浏览器

2，添加Header Editor浏览器插件【找到FireFox浏览器的 工具 -> 扩展和主题 —> 搜索 Header Editor -> 添加到FireFox】

3，配置Header Editor插件【找到FireFox右上角Header Editor -> 导出和导入 -> 下载规则中输入如下规则url -> 点击向下箭头加载 】

规则url: https://azurezeng.github.io/static/HE-GoogleRedirect.json

4，在kaggle官网用邮箱正常注册kaggle即可。

kaggle官网：https://www.kaggle.com/

5，此后就可以在任何能联网的地方正常登录kaggle，不再需要Header Editor了

二，设置GPU

1，新建notebook. 【点击kaggle主页面左上角+, 选择notebook】

2，开启GPU开关。【点击展开notebook右上角 |< 设置，设置Accelerator为GPU 】

3，查看GPU信息。【NoteBook中使用 nvidia-smi查看】

        
          
!nvidia-smi   

      

        
          
import torch   
torch.cuda.is_available() 
      

三，上传数据

1，点击展开notebook右上角 |< 设置，找到 Add Data，可以从Kaggle社区发布的数据集中选择一些想要的数据集。

2，也可以选在代表上传的向上箭头，上传数据集文件作为自定义数据集。建议压缩后上传，传输效率较高。

3，此外，也可以通过把数据放在github项目中，用git clone的方式获取。

4，notebook加载进来数据集后，可以在右边数据文件位置点击获取对应路径。

公众号 算法美食屋 回复关键词： pytorch ， 获取本范例所用数据集eat_pytorch_datasets百度云盘下载链接。

四，CPU训练代码

约14s一个Epoch

        
          
import torch   
from torch import nn  
from torch.utils.data import Dataset,DataLoader  
from torchvision import transforms as T  
from torchvision import datasets   
  
  
#======================================================================  
# 一，准备数据  
#======================================================================  
  
transform_img = T.Compose(  
    [T.ToTensor()])  
  
def transform\_label(x):  
    return torch.tensor(x)  
  
ds_train = datasets.ImageFolder("../input/eat-pytorch-datasets/eat\_pytorch\_datasets/cifar2/train/",  
            transform = transform_img,target_transform = transform_label)  
ds_val = datasets.ImageFolder("../input/eat-pytorch-datasets/eat\_pytorch\_datasets/cifar2/test/",  
            transform = transform_img,target_transform = transform_label)  
print(ds_train.class_to_idx)  
  
dl_train = DataLoader(ds_train,batch_size = 50,shuffle = True, pin_memory=True, num_workers = 8)  
dl_val = DataLoader(ds_val,batch_size = 50,shuffle = False, pin_memory=True, num_workers = 8)  
  
  
%matplotlib inline  
%config InlineBackend.figure_format = 'svg'  
  
#查看部分样本  
from matplotlib import pyplot as plt   
  
plt.figure(figsize=(8,8))   
for i in range(9):  
    img,label = ds_train[i]  
    img = img.permute(1,2,0)   
    ax=plt.subplot(3,3,i+1)  
    ax.imshow(img.numpy())  
    ax.set_title("label = %d"%label.item())  
    ax.set_xticks([])  
    ax.set_yticks([])   
plt.show()  
  
# Pytorch的图片默认顺序是 Batch,Channel,Width,Height  
for features,labels in dl_train:  
    print(features.shape,labels.shape)   
    break  
      
      
      
  
#======================================================================  
# 二，定义模型  
#======================================================================  
  
class Net(nn.Module):  
      
    def \_\_init\_\_(self):  
        super(Net, self).__init__()  
        self.conv1 = nn.Conv2d(in_channels=3,out_channels=32,kernel_size = 3)  
        self.pool = nn.MaxPool2d(kernel_size = 2,stride = 2)  
        self.conv2 = nn.Conv2d(in_channels=32,out_channels=64,kernel_size = 5)  
        self.dropout = nn.Dropout2d(p = 0.1)  
        self.adaptive_pool = nn.AdaptiveMaxPool2d((1,1))  
        self.flatten = nn.Flatten()  
        self.linear1 = nn.Linear(64,32)  
        self.relu = nn.ReLU()  
        self.linear2 = nn.Linear(32,2)  
          
    def forward(self,x):  
        x = self.conv1(x)  
        x = self.pool(x)  
        x = self.conv2(x)  
        x = self.pool(x)  
        x = self.dropout(x)  
        x = self.adaptive_pool(x)  
        x = self.flatten(x)  
        x = self.linear1(x)  
        x = self.relu(x)  
        x = self.linear2(x)  
        return x   
          
net = Net()  
print(net)  
  
  
  
  
#======================================================================  
# 三，训练模型(CPU)  
#======================================================================  
  
  
import os,sys,time  
import numpy as np  
import pandas as pd  
import datetime   
from tqdm import tqdm   
  
import torch  
from torch import nn   
from copy import deepcopy  
from torchmetrics import Accuracy  
#注：多分类使用torchmetrics中的评估指标，二分类使用torchkeras.metrics中的评估指标  
  
def printlog(info):  
    nowtime = datetime.datetime.now().strftime('%Y-%m-%d %H:%M:%S')  
    print("\n"+"=========="*8 + "%s"%nowtime)  
    print(str(info)+"\n")  
      
  
net = Net()  
  
loss_fn = nn.CrossEntropyLoss()  
optimizer= torch.optim.Adam(net.parameters(),lr = 0.01)     
metrics_dict = {"acc":Accuracy()}  
  
epochs = 5   
ckpt_path='checkpoint.pt'  
  
#early\_stopping相关设置  
monitor="val\_acc"  
patience=3  
mode="max"  
  
history = {}  
  
for epoch in range(1, epochs+1):  
    printlog("Epoch {0} / {1}".format(epoch, epochs))  
  
    # 1，train -------------------------------------------------    
    net.train()  
      
    total_loss,step = 0,0  
      
    loop = tqdm(enumerate(dl_train), total =len(dl_train))  
    train_metrics_dict = deepcopy(metrics_dict)   
      
    for i, batch in loop:   
          
        features,labels = batch  
        #forward  
        preds = net(features)  
        loss = loss_fn(preds,labels)  
          
        #backward  
        loss.backward()  
        optimizer.step()  
        optimizer.zero_grad()  
              
        #metrics  
        step_metrics = {"train\_"+name:metric_fn(preds, labels).item()   
                        for name,metric_fn in train_metrics_dict.items()}  
          
        step_log = dict({"train\_loss":loss.item()},**step_metrics)  
  
        total_loss += loss.item()  
          
        step+=1  
        if i!=len(dl_train)-1:  
            loop.set_postfix(**step_log)  
        else:  
            epoch_loss = total_loss/step  
            epoch_metrics = {"train\_"+name:metric_fn.compute().item()   
                             for name,metric_fn in train_metrics_dict.items()}  
            epoch_log = dict({"train\_loss":epoch_loss},**epoch_metrics)  
            loop.set_postfix(**epoch_log)  
  
            for name,metric_fn in train_metrics_dict.items():  
                metric_fn.reset()  
                  
    for name, metric in epoch_log.items():  
        history[name] = history.get(name, []) + [metric]  
          
  
    # 2，validate -------------------------------------------------  
    net.eval()  
      
    total_loss,step = 0,0  
    loop = tqdm(enumerate(dl_val), total =len(dl_val))  
      
    val_metrics_dict = deepcopy(metrics_dict)   
      
    with torch.no_grad():  
        for i, batch in loop:   
  
            features,labels = batch  
              
            #forward  
            preds = net(features)  
            loss = loss_fn(preds,labels)  
  
            #metrics  
            step_metrics = {"val\_"+name:metric_fn(preds, labels).item()   
                            for name,metric_fn in val_metrics_dict.items()}  
  
            step_log = dict({"val\_loss":loss.item()},**step_metrics)  
  
            total_loss += loss.item()  
            step+=1  
            if i!=len(dl_val)-1:  
                loop.set_postfix(**step_log)  
            else:  
                epoch_loss = (total_loss/step)  
                epoch_metrics = {"val\_"+name:metric_fn.compute().item()   
                                 for name,metric_fn in val_metrics_dict.items()}  
                epoch_log = dict({"val\_loss":epoch_loss},**epoch_metrics)  
                loop.set_postfix(**epoch_log)  
  
                for name,metric_fn in val_metrics_dict.items():  
                    metric_fn.reset()  
                      
    epoch_log["epoch"] = epoch             
    for name, metric in epoch_log.items():  
        history[name] = history.get(name, []) + [metric]  
  
    # 3，early-stopping -------------------------------------------------  
    arr_scores = history[monitor]  
    best_score_idx = np.argmax(arr_scores) if mode=="max" else np.argmin(arr_scores)  
    if best_score_idx==len(arr_scores)-1:  
        torch.save(net.state_dict(),ckpt_path)  
        print("<<<<<< reach best {0} : {1} >>>>>>".format(monitor,  
             arr_scores[best_score_idx]),file=sys.stderr)  
    if len(arr_scores)-best_score_idx>patience:  
        print("<<<<<< {} without improvement in {} epoch, early stopping >>>>>>".format(  
            monitor,patience),file=sys.stderr)  
        break   
    net.load_state_dict(torch.load(ckpt_path))  
      
dfhistory = pd.DataFrame(history)  
  
  
  

      

五，GPU训练代码

约8s一个Epoch.

        
          
import torch   
from torch import nn  
from torch.utils.data import Dataset,DataLoader  
from torchvision import transforms as T  
from torchvision import datasets   
  
  
#======================================================================  
# 一，准备数据  
#======================================================================  
  
transform_img = T.Compose(  
    [T.ToTensor()])  
  
def transform\_label(x):  
    return torch.tensor(x)  
  
ds_train = datasets.ImageFolder("../input/eat-pytorch-datasets/eat\_pytorch\_datasets/cifar2/train/",  
            transform = transform_img,target_transform = transform_label)  
ds_val = datasets.ImageFolder("../input/eat-pytorch-datasets/eat\_pytorch\_datasets/cifar2/test/",  
            transform = transform_img,target_transform = transform_label)  
print(ds_train.class_to_idx)  
  
dl_train = DataLoader(ds_train,batch_size = 50,shuffle = True, pin_memory=True, num_workers = 8)  
dl_val = DataLoader(ds_val,batch_size = 50,shuffle = False, pin_memory=True, num_workers = 8)  
  
  
  
%matplotlib inline  
%config InlineBackend.figure_format = 'svg'  
  
#查看部分样本  
from matplotlib import pyplot as plt   
  
plt.figure(figsize=(8,8))   
for i in range(9):  
    img,label = ds_train[i]  
    img = img.permute(1,2,0)   
    ax=plt.subplot(3,3,i+1)  
    ax.imshow(img.numpy())  
    ax.set_title("label = %d"%label.item())  
    ax.set_xticks([])  
    ax.set_yticks([])   
plt.show()  
  
# Pytorch的图片默认顺序是 Batch,Channel,Width,Height  
for features,labels in dl_train:  
    print(features.shape,labels.shape)   
    break  
      
      
      
  
#======================================================================  
# 二，定义模型  
#======================================================================  
  
class Net(nn.Module):  
      
    def \_\_init\_\_(self):  
        super(Net, self).__init__()  
        self.conv1 = nn.Conv2d(in_channels=3,out_channels=32,kernel_size = 3)  
        self.pool = nn.MaxPool2d(kernel_size = 2,stride = 2)  
        self.conv2 = nn.Conv2d(in_channels=32,out_channels=64,kernel_size = 5)  
        self.dropout = nn.Dropout2d(p = 0.1)  
        self.adaptive_pool = nn.AdaptiveMaxPool2d((1,1))  
        self.flatten = nn.Flatten()  
        self.linear1 = nn.Linear(64,32)  
        self.relu = nn.ReLU()  
        self.linear2 = nn.Linear(32,2)  
          
    def forward(self,x):  
        x = self.conv1(x)  
        x = self.pool(x)  
        x = self.conv2(x)  
        x = self.pool(x)  
        x = self.dropout(x)  
        x = self.adaptive_pool(x)  
        x = self.flatten(x)  
        x = self.linear1(x)  
        x = self.relu(x)  
        x = self.linear2(x)  
        return x   
          
net = Net()  
print(net)  
  
  
  
  
#======================================================================  
# 三，训练模型(CPU)  
#======================================================================  
  
  
import os,sys,time  
import numpy as np  
import pandas as pd  
import datetime   
from tqdm import tqdm   
  
import torch  
from torch import nn   
from copy import deepcopy  
from torchmetrics import Accuracy  
#注：多分类使用torchmetrics中的评估指标，二分类使用torchkeras.metrics中的评估指标  
  
def printlog(info):  
    nowtime = datetime.datetime.now().strftime('%Y-%m-%d %H:%M:%S')  
    print("\n"+"=========="*8 + "%s"%nowtime)  
    print(str(info)+"\n")  
      
  
net = Net()  
  
loss_fn = nn.CrossEntropyLoss()  
optimizer= torch.optim.Adam(net.parameters(),lr = 0.01)     
metrics_dict = {"acc":Accuracy()}  
  
  
#------------------------------移动模型到GPU上------------------------------  
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")  
net.to(device)  
loss_fn.to(device)  
for name,fn in metrics_dict.items():  
    fn.to(device)  
#-------------------------------------------------------------------------  
  
  
epochs = 5  
ckpt_path='checkpoint.pt'  
  
#early\_stopping相关设置  
monitor="val\_acc"  
patience=3  
mode="max"  
  
history = {}  
  
for epoch in range(1, epochs+1):  
    printlog("Epoch {0} / {1}".format(epoch, epochs))  
  
    # 1，train -------------------------------------------------    
    net.train()  
      
    total_loss,step = 0,0  
      
    loop = tqdm(enumerate(dl_train), total =len(dl_train))  
    train_metrics_dict = deepcopy(metrics_dict)   
      
    for i, batch in loop:   
          
        features,labels = batch  
          
        #------------------------------移动数据到GPU上------------------------------  
        features = features.to(device)  
        labels = labels.to(device)  
        #-------------------------------------------------------------------------  
          
        #forward  
        preds = net(features)  
        loss = loss_fn(preds,labels)  
          
        #backward  
        loss.backward()  
        optimizer.step()  
        optimizer.zero_grad()  
              
        #metrics  
        step_metrics = {"train\_"+name:metric_fn(preds, labels).item()   
                        for name,metric_fn in train_metrics_dict.items()}  
          
        step_log = dict({"train\_loss":loss.item()},**step_metrics)  
  
        total_loss += loss.item()  
          
        step+=1  
        if i!=len(dl_train)-1:  
            loop.set_postfix(**step_log)  
        else:  
            epoch_loss = total_loss/step  
            epoch_metrics = {"train\_"+name:metric_fn.compute().item()   
                             for name,metric_fn in train_metrics_dict.items()}  
            epoch_log = dict({"train\_loss":epoch_loss},**epoch_metrics)  
            loop.set_postfix(**epoch_log)  
  
            for name,metric_fn in train_metrics_dict.items():  
                metric_fn.reset()  
                  
    for name, metric in epoch_log.items():  
        history[name] = history.get(name, []) + [metric]  
          
  
    # 2，validate -------------------------------------------------  
    net.eval()  
      
    total_loss,step = 0,0  
    loop = tqdm(enumerate(dl_val), total =len(dl_val))  
      
    val_metrics_dict = deepcopy(metrics_dict)   
      
    with torch.no_grad():  
        for i, batch in loop:   
  
            features,labels = batch  
              
            #------------------------------移动数据到GPU上------------------------------  
            features = features.to(device)  
            labels = labels.to(device)  
            #-------------------------------------------------------------------------  
              
            #forward  
            preds = net(features)  
            loss = loss_fn(preds,labels)  
  
            #metrics  
            step_metrics = {"val\_"+name:metric_fn(preds, labels).item()   
                            for name,metric_fn in val_metrics_dict.items()}  
  
            step_log = dict({"val\_loss":loss.item()},**step_metrics)  
  
            total_loss += loss.item()  
            step+=1  
            if i!=len(dl_val)-1:  
                loop.set_postfix(**step_log)  
            else:  
                epoch_loss = (total_loss/step)  
                epoch_metrics = {"val\_"+name:metric_fn.compute().item()   
                                 for name,metric_fn in val_metrics_dict.items()}  
                epoch_log = dict({"val\_loss":epoch_loss},**epoch_metrics)  
                loop.set_postfix(**epoch_log)  
  
                for name,metric_fn in val_metrics_dict.items():  
                    metric_fn.reset()  
                      
    epoch_log["epoch"] = epoch             
    for name, metric in epoch_log.items():  
        history[name] = history.get(name, []) + [metric]  
  
    # 3，early-stopping -------------------------------------------------  
    arr_scores = history[monitor]  
    best_score_idx = np.argmax(arr_scores) if mode=="max" else np.argmin(arr_scores)  
    if best_score_idx==len(arr_scores)-1:  
        torch.save(net.state_dict(),ckpt_path)  
        print("<<<<<< reach best {0} : {1} >>>>>>".format(monitor,  
             arr_scores[best_score_idx]),file=sys.stderr)  
    if len(arr_scores)-best_score_idx>patience:  
        print("<<<<<< {} without improvement in {} epoch, early stopping >>>>>>".format(  
            monitor,patience),file=sys.stderr)  
        break   
    net.load_state_dict(torch.load(ckpt_path))  
      
dfhistory = pd.DataFrame(history)