| PyTorch | MindSpore |
```python
trans = torchvision.transforms.Resize(...)
mnist_train = torchvision.datasets.FashionMNIST(..., transforms=trans, ...)
```
|
```python
trans = mindspore.dataset.vision.Resize(...)
mnist_train = mindspore.dataset.FashionMnistDataset(...)
mnist_train = mnist_train.map(trans, ...)
```
|
| PyTorch | MindSpore |
```python
...
img_resize = torchvision.transforms.Resize(...)(input_ids)
img_resize = torchvision.transforms.ToTensor()(img_resize)
tmp_tensor = torch.tensor(np.ones_like(img_resize))
img_resize = torch.mul(img_resize, tmp_tensor)
img_resize = torchvision.transforms.Normalize(...)(img_resize)
...
```
|
```python
...
img_resize = mindspore.dataset.vision.Resize(...)(input_ids)
img_resize = mindspore.dataset.vision.ToTensor()(img_resize)
tmp_array = np.ones_like(img_resize)
img_resize = np.multiply(img_resize, tmp_array)
img_resize = mindspore.dataset.vision.Normalize(...)(img_resize)
...
```
|
| PyTorch | MindSpore |
```python
import numpy as np
import torch
from torch.utils.data import DataLoader
x = np.random.randint(0, 255, size=(20, 32, 32, 3))
tensor_x = torch.Tensor(x)
dataloader = DataLoader(tensor_x, batch_size=10)
for i, data in enumerate(dataloader):
print(i, data.shape)
```
运行结果：
```text
0 torch.Size([10, 32, 32, 3])
1 torch.Size([10, 32, 32, 3])
```
|
```python
import numpy as np
import mindspore.dataset as ds
x = np.random.randint(0, 255, size=(20, 32, 32, 3))
dataset = ds.GeneratorDataset(x, column_names=["data"])
dataset = dataset.batch(10, drop_remainder=True)
for data in dataset:
print(data[0].shape)
```
运行结果：
```text
(10, 32, 32, 3)
(10, 32, 32, 3)
```
|