更新readme

gcu_mnist_example/gpu_mnist_example/README.md

@@ -0,0 +1,50 @@
+# 如何在启智平台上进行模型训练—GCU示例
+
+## 1.启智集群和智算集群的GCU训练样例
+
+###### 启智集群的示例代码：
+
+- 训练示例请参考示例中[train_gcu.py](./train_gcu.py)的代码注释
+
+## 2. 在openi上获取数据集，预训练模型，输出路径
+
+安装openi包
+
+```
+pip install -U openi
+```
+
+使用openi包
+
+```
+#导入包
+from openi.context import prepare, upload_openi
+
+#初始化导入数据集和预训练模型到容器内
+openi_context = prepare()
+
+#获取数据集路径，预训练模型路径，输出路径
+dataset_path = openi_context.dataset_path
+pretrain_model_path = openi_context.pretrain_model_path
+output_path = openi_context.output_path
+
+#回传结果到openi
+upload_openi()
+```
+
+## 3.FAQ
+
+### 3.1 关于启智平台公共库[openi](https://openi.pcl.ac.cn/OpenIOSSG/openi-pypi)：
+
+主要使用的方法有以下几个：
+
+```
+prepare                    准备数据集，模型，输出路径
+upload_openi               将训练镜像的输出结果拷贝回启智平台
+```
+
+### 3.2 解决参数报错问题：
+
+请在代码中加入 `args, unknown = parser.parse_known_args()`，可忽略掉 `--ckpt_url`， `--multi_date_url`等参数报错问题
+
+## 对于示例代码有任何问题，欢迎在本项目中提issue。

gcu_mnist_example/gpu_mnist_example/model.py

@@ -0,0 +1,35 @@
+from torch.nn import Module
+from torch import nn
+
+
+class Model(Module):
+    def __init__(self):
+        super(Model, self).__init__()
+        self.conv1 = nn.Conv2d(1, 6, 5)
+        self.relu1 = nn.ReLU()
+        self.pool1 = nn.MaxPool2d(2)
+        self.conv2 = nn.Conv2d(6, 16, 5)
+        self.relu2 = nn.ReLU()
+        self.pool2 = nn.MaxPool2d(2)
+        self.fc1 = nn.Linear(256, 120)
+        self.relu3 = nn.ReLU()
+        self.fc2 = nn.Linear(120, 84)
+        self.relu4 = nn.ReLU()
+        self.fc3 = nn.Linear(84, 10)
+        self.relu5 = nn.ReLU()
+
+    def forward(self, x):
+        y = self.conv1(x)
+        y = self.relu1(y)
+        y = self.pool1(y)
+        y = self.conv2(y)
+        y = self.relu2(y)
+        y = self.pool2(y)
+        y = y.view(y.shape[0], -1)
+        y = self.fc1(y)
+        y = self.relu3(y)
+        y = self.fc2(y)
+        y = self.relu4(y)
+        y = self.fc3(y)
+        y = self.relu5(y)
+        return y

gcu_mnist_example/gpu_mnist_example/train_gcu.py

@@ -0,0 +1,166 @@
+#!/usr/bin/python
+#coding=utf-8
+'''
+If there are Chinese comments in the code，please add at the beginning：
+#!/usr/bin/python
+#coding=utf-8   
+
+示例选用的数据集是MnistDataset.zip
+数据集结构是：
+ MnistDataset.zip
+  ├── test
+  │   ├── MNIST/processed/test.pt
+  │   └── MNIST/processed/training.pt
+  │   ├── MNIST/raw/train-images-idx3-ubyte
+  │   └── MNIST/raw/train-labels-idx1-ubyte
+  │   ├── MNIST/raw/t10k-images-idx3-ubyte
+  │   └── MNIST/raw/t10k-labels-idx1-ubyte
+  ├── train
+  │   ├── MNIST/processed/test.pt
+  │   └── MNIST/processed/training.pt
+  │   ├── MNIST/raw/train-images-idx3-ubyte
+  │   └── MNIST/raw/train-labels-idx1-ubyte
+  │   ├── MNIST/raw/t10k-images-idx3-ubyte
+  │   └── MNIST/raw/t10k-labels-idx1-ubyte
+
+
+
+示例选用的预训练模型文件夹为MNIST_PytorchExample_GPU_test34_model_7f9j，模型文件为：mnist_epoch1_0.70.pkl         
+
+'''
+
+import os
+print("begin:")
+os.system("pip uninstall openi-test")
+os.system("pip install {}".format(os.getenv("OPENI_SDK_PATH")))
+import torch
+from model import Model
+import numpy as np
+from torchvision.datasets import mnist
+from torch.nn import CrossEntropyLoss
+from torch.optim import SGD
+from torch.utils.data import DataLoader
+from torchvision.transforms import ToTensor
+import argparse
+from openi.context import prepare, upload_openi
+
+import importlib.util
+
+def is_torch_dtu_available():
+    if importlib.util.find_spec("torch_dtu") is None:
+        return False
+    if importlib.util.find_spec("torch_dtu.core") is None:
+        return False
+    return importlib.util.find_spec("torch_dtu.core.dtu_model") is not None
+
+# Training settings
+parser = argparse.ArgumentParser(description='PyTorch MNIST Example')
+parser.add_argument('--epoch_size', type=int, default=1, help='how much epoch to train')
+parser.add_argument('--batch_size', type=int, default=256, help='how much batch_size in epoch')
+
+
+if __name__ == '__main__':
+    #获取参数并忽略超参数报错
+    args, unknown = parser.parse_known_args()
+    
+    #初始化导入数据集和预训练模型到容器内
+    openi_context = prepare()
+
+    #获取数据集路径，预训练模型路径，输出路径
+    dataset_path = openi_context.dataset_path
+    pretrain_model_path = openi_context.pretrain_model_path
+    output_path = openi_context.output_path
+
+    dataset_path_A = dataset_path + "/MnistDataset"
+    pretrain_model_path_A = pretrain_model_path + "/MNIST_PytorchExample_GPU_test34_model_7f9j"
+
+    print("dataset_path:")
+    print(os.listdir(dataset_path))
+    os.listdir(dataset_path)
+    
+    print("pretrain_model_path:")
+    print(os.listdir(pretrain_model_path))
+    os.listdir(pretrain_model_path)
+
+    print("output_path:")
+    print(os.listdir(output_path))
+    os.listdir(output_path)
+    # load DPU envs-xx.sh
+    DTU_FLAG = True
+    if is_torch_dtu_available():
+        import torch_dtu
+        import torch_dtu.distributed as dist
+        import torch_dtu.core.dtu_model as dm
+        from torch_dtu.nn.parallel import DistributedDataParallel as torchDDP
+        print('dtu is available: True') 
+        device = dm.dtu_device()
+        DTU_FLAG = True
+    else:
+        print('dtu is available: False')
+        device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
+        DTU_FLAG = False
+        
+    
+    # 参数声明
+    model = Model().to(device)
+    optimizer = SGD(model.parameters(), lr=1e-1)
+    #log output
+    batch_size = args.batch_size
+    train_dataset = mnist.MNIST(root=dataset_path_A + "/train", train=True, transform=ToTensor(),download=False)
+    test_dataset = mnist.MNIST(root=dataset_path_A + "/test", train=False, transform=ToTensor(),download=False)
+    train_loader = DataLoader(train_dataset, batch_size=batch_size)
+    test_loader = DataLoader(test_dataset, batch_size=batch_size)
+    model = Model().to(device)
+    sgd = SGD(model.parameters(), lr=1e-1)
+    cost = CrossEntropyLoss()
+    epochs = args.epoch_size
+    print('epoch_size is:{}'.format(epochs))
+
+    # 如果有保存的模型，则加载模型，并在其基础上继续训练
+    if os.path.exists(pretrain_model_path_A+"/mnist_epoch1_0.70.pkl"):
+        checkpoint = torch.load(pretrain_model_path_A+"/mnist_epoch1_0.70.pkl")
+        model.load_state_dict(checkpoint['model'])
+        optimizer.load_state_dict(checkpoint['optimizer'])
+        start_epoch = checkpoint['epoch']
+        print('加载 epoch {} 权重成功！'.format(start_epoch))
+    else:
+        start_epoch = 0
+        print('无保存模型，将从头开始训练！')
+
+    for _epoch in range(start_epoch, epochs):
+        print('the {} epoch_size begin'.format(_epoch + 1))
+        model.train()
+        for idx, (train_x, train_label) in enumerate(train_loader):
+            train_x = train_x.to(device)
+            train_label = train_label.to(device)
+            label_np = np.zeros((train_label.shape[0], 10))
+            sgd.zero_grad()
+            predict_y = model(train_x.float())
+            loss = cost(predict_y, train_label.long())
+            if idx % 10 == 0:
+                print('idx: {}, loss: {}'.format(idx, loss.sum().item()))
+            loss.backward()
+            if DTU_FLAG:
+                dm.optimizer_step(sgd, barrier=True)
+            else:
+                sgd.step()
+            
+            
+        correct = 0
+        _sum = 0
+        model.eval()
+        for idx, (test_x, test_label) in enumerate(test_loader):
+            test_x = test_x
+            test_label = test_label
+            predict_y = model(test_x.to(device).float()).detach()
+            predict_ys = np.argmax(predict_y.cpu(), axis=-1)
+            label_np = test_label.numpy()
+            _ = predict_ys == test_label
+            correct += np.sum(_.numpy(), axis=-1)
+            _sum += _.shape[0]
+        print('accuracy: {:.2f}'.format(correct / _sum))
+        #The model output location is placed under output_path
+        state = {'model':model.state_dict(), 'optimizer':optimizer.state_dict(), 'epoch':_epoch+1}
+        torch.save(state, '{}/mnist_epoch{}_{:.2f}.pkl'.format(output_path,_epoch+1, correct / _sum))
+        print('test:')
+        print(os.listdir(output_path))

gpu_mnist_example/README.md

@@ -4,7 +4,7 @@
 
 ###### 启智集群的示例代码：
 
-- 训练示例请参考示例中[train_npu.py](./train_npu.py)的代码注释
+- 训练示例请参考示例中[train_gpu.py](./train_gpu.py)的代码注释
 
 ## 2. 在openi上获取数据集，预训练模型，输出路径
 
@@ -43,8 +43,7 @@ upload_openi()
 ```
 prepare                    准备数据集，模型，输出路径
 upload_openi               将训练镜像的输出结果拷贝回启智平台
-obs_copy_file              通过mox拷贝文件
-obs_copy_folder 	   通过mox拷贝文件夹
+
 ```
 
 ### 4.2 解决参数报错问题：

npu_mnist_example/README.md

@@ -59,7 +59,7 @@ obs_copy_folder 	   通过mox拷贝文件夹
 
 ```
 if local_rank%8==0:
-    这里省略下载数据的代码...
+    这里省略下载数据的代码...(openi.context.prepare())
     f = open("/cache/download_input.txt", 'w')  
     f.close()
     try: