KaneAI - World’s First E2E Software Testing Agent.Book a Demo Join TestMu Conference

How to use outputs method in localstack

Best Python code snippet using localstack_python

models.py

Source:models.py

1import tensorflow as tf2from ..utils.tf_utils import activation_fun3def UNET(nb_classes, inputs):4    """Compile a UNET model.5    Args:6        nb_classes: the number of classes to predict7        inputs: the input tensor8    Returns:9        an output tensor, with 'nb_classes' of featuremaps10    """11    padding = 'same'12    # Conv block 113    outputs = tf.compat.v1.layers.conv2d(inputs, 64, 3, padding=padding, kernel_initializer=tf.compat.v1.keras.initializers.VarianceScaling(scale=1.0, mode="fan_avg", distribution="uniform"), name='conv1-1', use_bias=True)14    outputs = tf.nn.relu(outputs)15    16    outputs = tf.compat.v1.layers.conv2d(outputs, 64, 3, padding=padding, kernel_initializer=tf.compat.v1.keras.initializers.VarianceScaling(scale=1.0, mode="fan_avg", distribution="uniform"), name='conv1-2', use_bias=True)17    outputs = tf.nn.relu(outputs)18    # Make a copy of conv1 output tensor 19    conv1_output = outputs20    21    # Down-sample 122    outputs = tf.compat.v1.layers.max_pooling2d(outputs,pool_size = 2,strides = 2,padding=padding)23    24    # Conv block 225    outputs = tf.compat.v1.layers.conv2d(outputs, 128, 3, padding=padding, kernel_initializer=tf.compat.v1.keras.initializers.VarianceScaling(scale=1.0, mode="fan_avg", distribution="uniform"), name='conv2-1', use_bias=True)26    outputs = tf.nn.relu(outputs)27    outputs = tf.compat.v1.layers.conv2d(outputs, 128, 3, padding=padding, kernel_initializer=tf.compat.v1.keras.initializers.VarianceScaling(scale=1.0, mode="fan_avg", distribution="uniform"), name='conv2-2', use_bias=True)28    outputs = tf.nn.relu(outputs)29    # Make a copy of conv2 output tensor 30    conv2_output = outputs31    32    # Down-sample 233    outputs = tf.compat.v1.layers.max_pooling2d(outputs,pool_size = 2,strides = 2,padding=padding)34    35    # Conv block 336    outputs = tf.compat.v1.layers.conv2d(outputs, 256, 3, padding=padding, kernel_initializer=tf.compat.v1.keras.initializers.VarianceScaling(scale=1.0, mode="fan_avg", distribution="uniform"), name='conv3-1', use_bias=True)37    outputs = tf.nn.relu(outputs)38    outputs = tf.compat.v1.layers.conv2d(outputs, 256, 3, padding=padding, kernel_initializer=tf.compat.v1.keras.initializers.VarianceScaling(scale=1.0, mode="fan_avg", distribution="uniform"), name='conv3-2', use_bias=True)39    outputs = tf.nn.relu(outputs)40    # Make a copy of conv3 output tensor 41    conv3_output = outputs42    43    # Down-sample 344    outputs = tf.compat.v1.layers.max_pooling2d(outputs,pool_size = 2,strides = 2,padding=padding)45    46    # Conv block 447    outputs = tf.compat.v1.layers.conv2d(outputs, 512, 3, padding=padding, kernel_initializer=tf.compat.v1.keras.initializers.VarianceScaling(scale=1.0, mode="fan_avg", distribution="uniform"), name='conv4-1', use_bias=True)48    outputs = tf.nn.relu(outputs)49    outputs = tf.compat.v1.layers.conv2d(outputs, 512, 3, padding=padding, kernel_initializer=tf.compat.v1.keras.initializers.VarianceScaling(scale=1.0, mode="fan_avg", distribution="uniform"), name='conv4-2', use_bias=True)50    outputs = tf.nn.relu(outputs)51    # Make a copy of conv4 output tensor 52    conv4_output = outputs53    54    # Down-sample 455    outputs = tf.compat.v1.layers.max_pooling2d(outputs,pool_size = 2,strides = 2,padding=padding)56    57    # Get extracted feature for RPN58    rpn_feature = outputs59    60    61    # Conv block 562    outputs = tf.compat.v1.layers.conv2d(outputs, 1024, 3, padding=padding, kernel_initializer=tf.compat.v1.keras.initializers.VarianceScaling(scale=1.0, mode="fan_avg", distribution="uniform"), name='conv5-1', use_bias=True)63    outputs = tf.nn.relu(outputs)64    outputs = tf.compat.v1.layers.conv2d(outputs, 1024, 3, padding=padding, kernel_initializer=tf.compat.v1.keras.initializers.VarianceScaling(scale=1.0, mode="fan_avg", distribution="uniform"), name='conv5-2', use_bias=True)65    outputs = tf.nn.relu(outputs)66    67    68    # Up-sample(Conv_transpose) 469    outputs = tf.compat.v1.layers.conv2d_transpose(outputs, 512, 3, strides=(2, 2),70 padding=padding, kernel_initializer=tf.compat.v1.keras.initializers.VarianceScaling(scale=1.0, mode="fan_avg", distribution="uniform"), use_bias=True)71    outputs = tf.nn.relu(outputs)72    73    # changing the former line to the line below will connect the 4th layer on both ends, 74    # and form the classic U-Net architecture, but in our experiments we found this did 75    # not gain a better performance, also thanks for Meryem Uzun-Per for pointing this out76    77    # outputs = tf.concat([conv4_output, outputs], 3)78    79    # Conv block 4'80    outputs = tf.compat.v1.layers.conv2d(outputs, 512, 3, padding=padding, kernel_initializer=tf.compat.v1.keras.initializers.VarianceScaling(scale=1.0, mode="fan_avg", distribution="uniform"), name='conv4-3', use_bias=True)81    outputs = tf.nn.relu(outputs)82    outputs = tf.compat.v1.layers.conv2d(outputs, 512, 3, padding=padding, kernel_initializer=tf.compat.v1.keras.initializers.VarianceScaling(scale=1.0, mode="fan_avg", distribution="uniform"), name='conv4-4', use_bias=True)83    outputs = tf.nn.relu(outputs)84    85    # Up-sample(Conv_transpose) 386    outputs = tf.compat.v1.layers.conv2d_transpose(outputs, 256, 3, strides=(2, 2),87 padding=padding, kernel_initializer=tf.compat.v1.keras.initializers.VarianceScaling(scale=1.0, mode="fan_avg", distribution="uniform"), use_bias=True)88    outputs = tf.concat([conv3_output, outputs], 3)89    90    # Conv block 3'91    outputs = tf.compat.v1.layers.conv2d(outputs, 256, 3, padding=padding, kernel_initializer=tf.compat.v1.keras.initializers.VarianceScaling(scale=1.0, mode="fan_avg", distribution="uniform"), name='conv3-3', use_bias=True)92    outputs = tf.nn.relu(outputs)93    outputs = tf.compat.v1.layers.conv2d(outputs, 256, 3, padding=padding, kernel_initializer=tf.compat.v1.keras.initializers.VarianceScaling(scale=1.0, mode="fan_avg", distribution="uniform"), name='conv3-4', use_bias=True)94    outputs = tf.nn.relu(outputs)95    96    # Up-sample(Conv_transpose) 297    outputs = tf.compat.v1.layers.conv2d_transpose(outputs, 128, 3, strides=(2, 2),98 padding=padding, kernel_initializer=tf.compat.v1.keras.initializers.VarianceScaling(scale=1.0, mode="fan_avg", distribution="uniform"), use_bias=True)99    outputs = tf.concat([conv2_output, outputs], 3)100    101    # Conv block 2'102    outputs = tf.compat.v1.layers.conv2d(outputs, 128, 3, padding=padding, kernel_initializer=tf.compat.v1.keras.initializers.VarianceScaling(scale=1.0, mode="fan_avg", distribution="uniform"), name='conv2-3', use_bias=True)103    outputs = tf.nn.relu(outputs)104    outputs = tf.compat.v1.layers.conv2d(outputs, 128, 3, padding=padding, kernel_initializer=tf.compat.v1.keras.initializers.VarianceScaling(scale=1.0, mode="fan_avg", distribution="uniform"), name='conv2-4', use_bias=True)105    outputs = tf.nn.relu(outputs)106    107    # Up-sample(Conv_transpose) 1108    outputs = tf.compat.v1.layers.conv2d_transpose(outputs, 64, 3, strides=(2, 2),109 padding=padding, kernel_initializer=tf.compat.v1.keras.initializers.VarianceScaling(scale=1.0, mode="fan_avg", distribution="uniform"), use_bias=True)110    outputs = tf.concat([conv1_output, outputs], 3)111    112    # Conv block 2'113    outputs = tf.compat.v1.layers.conv2d(outputs, 64, 3, padding=padding, kernel_initializer=tf.compat.v1.keras.initializers.VarianceScaling(scale=1.0, mode="fan_avg", distribution="uniform"), name='conv1-3', use_bias=True)114    outputs = tf.nn.relu(outputs)115    outputs = tf.compat.v1.layers.conv2d(outputs, 64, 3, padding=padding, kernel_initializer=tf.compat.v1.keras.initializers.VarianceScaling(scale=1.0, mode="fan_avg", distribution="uniform"), name='conv1-4', use_bias=True)116    outputs = tf.nn.relu(outputs)117    118    # only output 2 featuremaps at the end119    outputs = tf.compat.v1.layers.conv2d(outputs, nb_classes, 3, padding=padding, kernel_initializer=tf.compat.v1.keras.initializers.VarianceScaling(scale=1.0, mode="fan_avg", distribution="uniform"), name='final', use_bias=False)120    ...

DCGAN.py

Source:DCGAN.py

1import torch2import torch.nn as nn3class Generator(nn.Module):4    def __init__(self):5        super(Generator, self).__init__()6        self.encoder_conv_1 = nn.Sequential(7            nn.Conv2d(1, 64, kernel_size=3, stride=1, padding=1),8            nn.LeakyReLU(inplace=False),9        )10        self.encoder_conv_2 = nn.Sequential(11            nn.Conv2d(64, 128, kernel_size=4, stride=2, padding=1),12            nn.LeakyReLU(inplace=False),13        )14        self.encoder_conv_3 = nn.Sequential(15            nn.Conv2d(128, 256, kernel_size=4, stride=2, padding=1),16            nn.LeakyReLU(inplace=False),17        )18        self.encoder_conv_4 = nn.Sequential(19            nn.Conv2d(256, 512, kernel_size=4, stride=2, padding=1),20            nn.LeakyReLU(inplace=False),21        )22        self.encoder_conv_5 = nn.Sequential(23            nn.Conv2d(512, 512, kernel_size=4, stride=2, padding=1),24            nn.LeakyReLU(inplace=False),25        )26        self.encoder_conv_6 = nn.Sequential(27            nn.Conv2d(512, 512, kernel_size=4, stride=2, padding=1),28            nn.LeakyReLU(inplace=False),29        )30        self.decoder_conv_1 = nn.Sequential(31            nn.ConvTranspose2d(512, 512, kernel_size=4, stride=2, padding=1),32            nn.ReLU(inplace=False),33            nn.Dropout(0.5),34        )35        self.decoder_conv_2 = nn.Sequential(36            nn.ConvTranspose2d(1024, 512, kernel_size=4, stride=2, padding=1),37            nn.ReLU(inplace=False),38            nn.Dropout(0.5),39        )40        self.decoder_conv_3 = nn.Sequential(41            nn.ConvTranspose2d(1024, 256, kernel_size=4, stride=2, padding=1),42            nn.ReLU(inplace=False),43        )44        self.decoder_conv_4 = nn.Sequential(45            nn.ConvTranspose2d(512, 128, kernel_size=4, stride=2, padding=1),46            nn.ReLU(inplace=False),47        )48        self.decoder_conv_5 = nn.Sequential(49            nn.ConvTranspose2d(256, 64, kernel_size=4, stride=2, padding=1),50            nn.ReLU(inplace=False),51        )52        self.output = nn.Sequential(53            nn.Conv2d(64, 3, kernel_size=1, stride=1, padding=0),54            nn.Tanh(),55        )56    def forward(self, inputs):57        encoder_conv_1_outputs = self.encoder_conv_1(inputs)58        encoder_conv_2_outputs = self.encoder_conv_2(encoder_conv_1_outputs)59        encoder_conv_3_outputs = self.encoder_conv_3(encoder_conv_2_outputs)60        encoder_conv_4_outputs = self.encoder_conv_4(encoder_conv_3_outputs)61        encoder_conv_5_outputs = self.encoder_conv_5(encoder_conv_4_outputs)62        encoder_conv_6_outputs = self.encoder_conv_6(encoder_conv_5_outputs)63        decoder_conv_1_outputs = self.decoder_conv_1(encoder_conv_6_outputs)64        decoder_conv_2_outputs = self.decoder_conv_2(torch.cat([decoder_conv_1_outputs, encoder_conv_5_outputs], dim=1))65        decoder_conv_3_outputs = self.decoder_conv_3(torch.cat([decoder_conv_2_outputs, encoder_conv_4_outputs], dim=1))66        decoder_conv_4_outputs = self.decoder_conv_4(torch.cat([decoder_conv_3_outputs, encoder_conv_3_outputs], dim=1))67        decoder_conv_5_outputs = self.decoder_conv_5(torch.cat([decoder_conv_4_outputs, encoder_conv_2_outputs], dim=1))68        outputs = self.output(decoder_conv_5_outputs)69        return outputs70class Discriminator(nn.Module):71    def __init__(self):72        super(Discriminator, self).__init__()73        self.conv1 = nn.Sequential(74            nn.Conv2d(4, 64, kernel_size=4, stride=2, padding=1),75            nn.LeakyReLU(inplace=False),76            nn.BatchNorm2d(64),77        )78        self.conv2 = nn.Sequential(79            nn.Conv2d(64, 128, kernel_size=4, stride=2, padding=1),80            nn.LeakyReLU(inplace=False),81            nn.BatchNorm2d(128),82        )83        self.conv3 = nn.Sequential(84            nn.Conv2d(128, 256, kernel_size=4, stride=2, padding=1),85            nn.LeakyReLU(inplace=False),86            nn.BatchNorm2d(256),87        )88        self.conv4 = nn.Sequential(89            nn.Conv2d(256, 512, kernel_size=4, stride=2, padding=1),90            nn.LeakyReLU(inplace=False),91            nn.BatchNorm2d(512),92        )93        self.conv5 = nn.Sequential(94            nn.Conv2d(512, 512, kernel_size=3, stride=1, padding=1),95            nn.LeakyReLU(inplace=False),96            nn.BatchNorm2d(512),97        )98        self.output = nn.Conv2d(512, 1, kernel_size=4, stride=1, padding=0)99    def forward(self, inputs):100        outputs = self.conv1(inputs)101        outputs = self.conv2(outputs)102        outputs = self.conv3(outputs)103        outputs = self.conv4(outputs)104        outputs = self.conv5(outputs)105        outputs = self.output(outputs)...

test_cleaner.py

Source:test_cleaner.py

1# coding: utf-82import os3import shutil4from abiflows.core.mastermind_abc import Cleaner5from abiflows.core.testing import AbiflowsTest6test_dir = os.path.join(os.path.dirname(__file__), "..", "..", "..", "..",7                        "test_files")8class TestCleaner(AbiflowsTest):9    def test_cleaner(self):10        # Keep current working directory, create tmp directory and change to tmp directory11        cwd = os.getcwd()12        tmp_dir = '_tmp_cleaner'13        if os.path.exists(tmp_dir):14            shutil.rmtree(tmp_dir)15        os.makedirs(tmp_dir)16        os.chdir(tmp_dir)17        tmp_abs_dir = os.getcwd()18        # Create a list of files and directories19        os.makedirs('outputs')20        os.makedirs('outputs/formatted')21        os.makedirs('outputs/text')22        os.makedirs('results')23        os.makedirs('temporary')24        open('somefile.txt', "w").close()25        open('somefile.txt.backup', "w").close()26        open('outputs/text/text1.abc', "w").close()27        open('outputs/text/text2.abc', "w").close()28        open('outputs/text/text3.abc', "w").close()29        open('outputs/text/text1.def', "w").close()30        open('outputs/text/text15.def', "w").close()31        open('outputs/formatted/formatted1.txt', "w").close()32        open('outputs/formatted/formatted2.txt', "w").close()33        open('outputs/formatted/formatted3.log', "w").close()34        open('outputs/formatted/formatted4.log', "w").close()35        open('outputs/formatted/formatted5.log', "w").close()36        open('outputs/formatted/formatted6.bin', "w").close()37        open('outputs/formatted/formatted7.bog', "w").close()38        open('outputs/formatted/formatted8.beg', "w").close()39        open('temporary/item.log', "w").close()40        open('temporary/result.txt', "w").close()41        # Create a first cleaner42        cleaner1 = Cleaner(dirs_and_patterns=[{'directory': 'outputs/text',43                                               'patterns': ['text?.abc']}])44        cleaner1.clean(root_directory=tmp_abs_dir)45        # Check that the first cleaner did his job correctly46        self.assertTrue(os.path.exists('outputs/text/text1.def'))47        self.assertTrue(os.path.exists('outputs/text/text15.def'))48        self.assertFalse(os.path.exists('outputs/text/text1.abc'))49        self.assertFalse(os.path.exists('outputs/text/text2.abc'))50        self.assertFalse(os.path.exists('outputs/text/text3.abc'))51        # Create a second cleaner52        cleaner2 = Cleaner(dirs_and_patterns=[{'directory': '.',53                                               'patterns': ['temporary']},54                                              {'directory': 'outputs/formatted',55                                               'patterns': ['*[1-4].log', '*.b?g']}])56        cleaner2.clean(root_directory=tmp_abs_dir)57        # Check that the first cleaner did his job correctly58        self.assertTrue(os.path.exists('outputs/formatted/formatted1.txt'))59        self.assertTrue(os.path.exists('outputs/formatted/formatted2.txt'))60        self.assertFalse(os.path.exists('outputs/formatted/formatted3.log'))61        self.assertFalse(os.path.exists('outputs/formatted/formatted4.log'))62        self.assertTrue(os.path.exists('outputs/formatted/formatted5.log'))63        self.assertTrue(os.path.exists('outputs/formatted/formatted6.bin'))64        self.assertFalse(os.path.exists('outputs/formatted/formatted7.bog'))65        self.assertFalse(os.path.exists('outputs/formatted/formatted8.beg'))66        self.assertFalse(os.path.exists('temporary'))67        self.assertTrue(os.path.exists('outputs/formatted'))68        # Change back to the initial working directory and remove the tmp directory69        os.chdir(cwd)...

Automation Testing Tutorials

Learn to execute automation testing from scratch with LambdaTest Learning Hub. Right from setting up the prerequisites to run your first automation test, to following best practices and diving deeper into advanced test scenarios. LambdaTest Learning Hubs compile a list of step-by-step guides to help you be proficient with different test automation frameworks i.e. Selenium, Cypress, TestNG etc.