add src dir

model_zoo/alexnet/README.md

@@ -2,7 +2,7 @@
 
 ## Description
 
-Training AlexNet with CIFAR-10 dataset in MindSpore.
+Training AlexNet with dataset in MindSpore.
 
 This is the simple tutorial for training AlexNet in MindSpore.
 
@@ -10,19 +10,19 @@ This is the simple tutorial for training AlexNet in MindSpore.
 
 - Install [MindSpore](https://www.mindspore.cn/install/en).
 
-- Download the CIFAR-10 dataset, the directory structure is as follows:
+- Download the dataset, the directory structure is as follows:
 
 ```
-├─cifar-10-batches-bin
+├─10-batches-bin
 │
-└─cifar-10-verify-bin
+└─10-verify-bin
 ```
 
 ## Running the example
 
 ```python
 # train AlexNet, hyperparameter setting in config.py
-python train.py --data_path cifar-10-batches-bin
+python train.py --data_path 10-batches-bin
 ```
 
 You will get the loss value of each step as following:
@@ -38,8 +38,8 @@ epoch: 1 step: 1538, loss is 1.0221305
 
 Then, evaluate AlexNet according to network model
 ```python
-# evaluate AlexNet, 1 epoch training accuracy is up to 51.1%; 10 epoch training accuracy is up to 81.2%
+# evaluate AlexNet
-python eval.py --data_path cifar-10-verify-bin --ckpt_path checkpoint_alexnet-1_1562.ckpt
+python eval.py --data_path 10-verify-bin --ckpt_path checkpoint_alexnet-1_1562.ckpt
 ```
 
 ## Note

model_zoo/alexnet/eval.py

@@ -19,9 +19,9 @@ python eval.py --data_path /YourDataPath --ckpt_path Your.ckpt
 """
 
 import argparse
-from config import alexnet_cfg as cfg
+from src.config import alexnet_cfg as cfg
-from dataset import create_dataset_mnist
+from src.dataset import create_dataset_mnist
-from alexnet import AlexNet
+from src.alexnet import AlexNet
 import mindspore.nn as nn
 from mindspore import context
 from mindspore.train.serialization import load_checkpoint, load_param_into_net

model_zoo/alexnet/train.py

@@ -19,10 +19,10 @@ python train.py --data_path /YourDataPath
 """
 
 import argparse
-from config import alexnet_cfg as cfg
+from src.config import alexnet_cfg as cfg
-from dataset import create_dataset_mnist
+from src.dataset import create_dataset_mnist
-from generator_lr import get_lr
+from src.generator_lr import get_lr
-from alexnet import AlexNet
+from src.alexnet import AlexNet
 import mindspore.nn as nn
 from mindspore import context
 from mindspore import Tensor

model_zoo/lenet/README.md

@@ -2,7 +2,7 @@
 
 ## Description
 
-Training LeNet with MNIST dataset in MindSpore.
+Training LeNet with dataset in MindSpore.
 
 This is the simple and basic tutorial for constructing a network in MindSpore.
 
@@ -10,10 +10,10 @@ This is the simple and basic tutorial for constructing a network in MindSpore.
 
 - Install [MindSpore](https://www.mindspore.cn/install/en).
 
-- Download the MNIST dataset, the directory structure is as follows:
+- Download the dataset, the directory structure is as follows:
 
 ```
-└─MNIST_Data
+└─Data
     ├─test
     │      t10k-images.idx3-ubyte
     │      t10k-labels.idx1-ubyte
@@ -27,7 +27,7 @@ This is the simple and basic tutorial for constructing a network in MindSpore.
 
 ```python
 # train LeNet, hyperparameter setting in config.py
-python train.py --data_path MNIST_Data
+python train.py --data_path Data
 ```
 
 You will get the loss value of each step as following:
@@ -43,8 +43,8 @@ epoch: 1 step: 1741, loss is 0.05018193
 
 Then, evaluate LeNet according to network model
 ```python
-# evaluate LeNet, after 1 epoch training, the accuracy is up to 96.5%
+# evaluate LeNet
-python eval.py --data_path MNIST_Data --ckpt_path checkpoint_lenet-1_1875.ckpt
+python eval.py --data_path Data --ckpt_path checkpoint_lenet-1_1875.ckpt
 ```
 
 ## Note

model_zoo/lenet/eval.py

@@ -20,9 +20,9 @@ python eval.py --data_path /YourDataPath --ckpt_path Your.ckpt
 
 import os
 import argparse
-from dataset import create_dataset
+from src.dataset import create_dataset
-from config import mnist_cfg as cfg
+from src.config import mnist_cfg as cfg
-from lenet import LeNet5
+from src.lenet import LeNet5
 import mindspore.nn as nn
 from mindspore import context
 from mindspore.train.serialization import load_checkpoint, load_param_into_net
@@ -32,10 +32,10 @@ from mindspore.nn.metrics import Accuracy
 
 
 if __name__ == "__main__":
-    parser = argparse.ArgumentParser(description='MindSpore MNIST Example')
+    parser = argparse.ArgumentParser(description='MindSpore Lenet Example')
     parser.add_argument('--device_target', type=str, default="Ascend", choices=['Ascend', 'GPU', 'CPU'],
                         help='device where the code will be implemented (default: Ascend)')
-    parser.add_argument('--data_path', type=str, default="./MNIST_Data",
+    parser.add_argument('--data_path', type=str, default="./Data",
                         help='path where the dataset is saved')
     parser.add_argument('--ckpt_path', type=str, default="", help='if mode is test, must provide\
                         path where the trained ckpt file')

model_zoo/lenet/train.py

@@ -20,9 +20,9 @@ python train.py --data_path /YourDataPath
 
 import os
 import argparse
-from config import mnist_cfg as cfg
+from src.config import mnist_cfg as cfg
-from dataset import create_dataset
+from src.dataset import create_dataset
-from lenet import LeNet5
+from src.lenet import LeNet5
 import mindspore.nn as nn
 from mindspore import context
 from mindspore.train.callback import ModelCheckpoint, CheckpointConfig, LossMonitor, TimeMonitor
@@ -31,10 +31,10 @@ from mindspore.nn.metrics import Accuracy
 
 
 if __name__ == "__main__":
-    parser = argparse.ArgumentParser(description='MindSpore MNIST Example')
+    parser = argparse.ArgumentParser(description='MindSpore Lenet Example')
     parser.add_argument('--device_target', type=str, default="Ascend", choices=['Ascend', 'GPU', 'CPU'],
                         help='device where the code will be implemented (default: Ascend)')
-    parser.add_argument('--data_path', type=str, default="./MNIST_Data",
+    parser.add_argument('--data_path', type=str, default="./Data",
                         help='path where the dataset is saved')
     parser.add_argument('--dataset_sink_mode', type=bool, default=True, help='dataset_sink_mode is False or True')