update examples

Author: hanjr92

examples/basic_tutorials/cifar10_cnn.py

@@ -2,12 +2,12 @@
 # -*- coding: utf-8 -*-
 
 import os
-# os.environ['TL_BACKEND'] = 'paddle'
-os.environ['TL_BACKEND'] = 'tensorflow'
+os.environ['TL_BACKEND'] = 'paddle'
+# os.environ['TL_BACKEND'] = 'tensorflow'
 # os.environ['TL_BACKEND'] = 'mindspore'
 
 import time
-from tensorlayerx.dataflow import Dataset, Dataloader
+from tensorlayerx.dataflow import Dataset, DataLoader
 from tensorlayerx.vision.transforms import (
     Compose, Resize, RandomFlipHorizontal, RandomContrast, RandomBrightness, StandardizePerImage, RandomCrop
 )
@@ -110,11 +110,8 @@ def __len__(self):
 train_dataset = make_dataset(data=X_train, label=y_train, transforms=train_transforms)
 test_dataset = make_dataset(data=X_test, label=y_test, transforms=test_transforms)
 
-train_dataset = tlx.dataflow.FromGenerator(train_dataset, output_types=(tlx.float32, tlx.int64))
-test_dataset = tlx.dataflow.FromGenerator(test_dataset, output_types=(tlx.float32, tlx.int64))
-
-train_dataset = Dataloader(train_dataset, batch_size=batch_size, shuffle=True, shuffle_buffer_size=128)
-test_dataset = Dataloader(test_dataset, batch_size=batch_size)
+train_dataset = DataLoader(train_dataset, batch_size=batch_size, shuffle=True)
+test_dataset = DataLoader(test_dataset, batch_size=batch_size)
 
 
 class WithLoss(Module):

examples/basic_tutorials/cifar10_cnn_mixed_mindspore.py

@@ -12,7 +12,7 @@
 from tensorlayerx.vision.transforms import (
     Compose, Resize, RandomFlipHorizontal, RandomContrast, RandomBrightness, StandardizePerImage, RandomCrop, HWC2CHW
 )
-from tensorlayerx.dataflow import Dataset, Dataloader
+from tensorlayerx.dataflow import Dataset, DataLoader
 from mindspore.nn import WithLossCell, Adam
 from mindspore import ParameterTuple
 import mindspore.nn as nn
@@ -99,15 +99,8 @@ def __len__(self):
 train_dataset = make_dataset(data=X_train, label=y_train, transforms=train_transforms)
 test_dataset = make_dataset(data=X_test, label=y_test, transforms=test_transforms)
 
-train_dataset = tlx.dataflow.FromGenerator(
-    train_dataset, output_types=(tlx.float32, tlx.int64), column_names=['data', 'label']
-)
-test_dataset = tlx.dataflow.FromGenerator(
-    test_dataset, output_types=(tlx.float32, tlx.int64), column_names=['data', 'label']
-)
-
-train_dataset = Dataloader(train_dataset, batch_size=batch_size, shuffle=True, shuffle_buffer_size=128)
-test_dataset = Dataloader(test_dataset, batch_size=batch_size)
+train_dataset = DataLoader(train_dataset, batch_size=batch_size, shuffle=True)
+test_dataset = DataLoader(test_dataset, batch_size=batch_size)
 
 
 class GradWrap(Module):

examples/basic_tutorials/cifar10_cnn_mixed_paddle.py

@@ -8,7 +8,7 @@
 import paddle as pd
 from tensorlayerx.nn import Module
 import tensorlayerx as tlx
-from tensorlayerx.dataflow import Dataset, Dataloader
+from tensorlayerx.dataflow import Dataset, DataLoader
 from tensorlayerx.nn import (Conv2d, Dense, Flatten, MaxPool2d, BatchNorm2d)
 from tensorlayerx.vision.transforms import (
     Compose, Resize, RandomFlipHorizontal, RandomContrast, RandomBrightness, StandardizePerImage, RandomCrop
@@ -108,11 +108,8 @@ def __len__(self):
 train_dataset = make_dataset(data=X_train, label=y_train, transforms=train_transforms)
 test_dataset = make_dataset(data=X_test, label=y_test, transforms=test_transforms)
 
-train_dataset = tlx.dataflow.FromGenerator(train_dataset, output_types=(tlx.float32, tlx.int64))
-test_dataset = tlx.dataflow.FromGenerator(test_dataset, output_types=(tlx.float32, tlx.int64))
-
-train_dataset = Dataloader(train_dataset, batch_size=batch_size, shuffle=True, shuffle_buffer_size=128)
-test_dataset = Dataloader(test_dataset, batch_size=batch_size)
+train_dataset = DataLoader(train_dataset, batch_size=batch_size, shuffle=True)
+test_dataset = DataLoader(test_dataset, batch_size=batch_size)
 
 for epoch in range(n_epoch):
     train_loss, train_acc, n_iter = 0, 0, 0

examples/basic_tutorials/imdb_LSTM_simple.py

@@ -60,10 +60,7 @@ def forward(self, x):
 print_freq = 2
 
 train_dataset = imdbdataset(X=X_train, y=y_train)
-train_dataset = tlx.dataflow.FromGenerator(
-    train_dataset, output_types=[tlx.int64, tlx.int64], column_names=['data', 'label']
-)
-train_loader = tlx.dataflow.Dataloader(train_dataset, batch_size=batch_size, shuffle=True)
+train_loader = tlx.dataflow.DataLoader(train_dataset, batch_size=batch_size, shuffle=True)
 
 net = ImdbNet()
 train_weights = net.trainable_weights

examples/basic_tutorials/mnist_dataflow.py

@@ -75,10 +75,7 @@ def forward(self, x):
 
 
 train_dataset = mnistdataset1(data=X_train, label=y_train, transform=transform)
-train_dataset = tlx.dataflow.FromGenerator(
-    train_dataset, output_types=[tlx.float32, tlx.int64], column_names=['data', 'label']
-)
-train_loader = tlx.dataflow.Dataloader(train_dataset, batch_size=128, shuffle=False)
+train_loader = tlx.dataflow.DataLoader(train_dataset, batch_size=128, shuffle=False)
 
 for i in train_loader:
     print(i[0].shape, i[1])

examples/basic_tutorials/mnist_gan.py

@@ -10,7 +10,7 @@
 import numpy as np
 import tensorlayerx as tlx
 from tensorlayerx.nn import Module, Dense
-from tensorlayerx.dataflow import Dataset
+from tensorlayerx.dataflow import Dataset, DataLoader
 from tensorlayerx.model import TrainOneStep
 
 X_train, y_train, X_val, y_val, X_test, y_test = tlx.files.load_mnist_dataset(shape=(-1, 784))
@@ -33,10 +33,7 @@ def __len__(self):
 
 batch_size = 128
 train_dataset = mnistdataset(data=X_train, label=y_train)
-train_dataset = tlx.dataflow.FromGenerator(
-    train_dataset, output_types=[tlx.float32, tlx.int64], column_names=['data', 'label']
-)
-train_loader = tlx.dataflow.Dataloader(train_dataset, batch_size=batch_size, shuffle=True)
+train_loader = tlx.dataflow.DataLoader(train_dataset, batch_size=batch_size, shuffle=True)
 
 
 class generator(Module):

examples/basic_tutorials/mnist_mlp.py

@@ -10,7 +10,7 @@
 import tensorlayerx as tlx
 from tensorlayerx.nn import Module
 from tensorlayerx.nn import Dense, Dropout
-from tensorlayerx.dataflow import Dataset
+from tensorlayerx.dataflow import Dataset, DataLoader
 
 X_train, y_train, X_val, y_val, X_test, y_test = tlx.files.load_mnist_dataset(shape=(-1, 784))
 
@@ -63,12 +63,11 @@ def forward(self, x, foo=None):
 metric = tlx.metrics.Accuracy()
 loss_fn = tlx.losses.softmax_cross_entropy_with_logits
 train_dataset = mnistdataset(data=X_train, label=y_train)
-train_dataset = tlx.dataflow.FromGenerator(
-    train_dataset, output_types=[tlx.float32, tlx.int64], column_names=['data', 'label']
-)
-train_loader = tlx.dataflow.Dataloader(train_dataset, batch_size=batch_size, shuffle=True)
+train_loader = tlx.dataflow.DataLoader(train_dataset, batch_size=batch_size, shuffle=True)
 
 model = tlx.model.Model(network=MLP, loss_fn=loss_fn, optimizer=optimizer, metrics=metric)
 model.train(n_epoch=n_epoch, train_dataset=train_loader, print_freq=print_freq, print_train_batch=False)
 model.save_weights('./model.npz', format='npz_dict')
 model.load_weights('./model.npz', format='npz_dict')
+
+