Added Test Cases

tests/test_conv_skip_connection.py

@@ -0,0 +1,16 @@
+import sys, os
+
+sys.path.append(os.path.dirname(os.getcwd()))
+
+from tensorflow import keras
+import numpy as np
+from pyradox import modules
+
+
+def test():
+    inputs = keras.Input(shape=(28, 28, 1))
+    x = modules.ConvSkipConnection(32, batch_normalization=True, dropout=0.2)(inputs)
+    x = keras.layers.GlobalAvgPool2D()(x)
+    outputs = keras.layers.Dense(10, activation="softmax")(x)
+
+    model = keras.models.Model(inputs=inputs, outputs=outputs)

tests/test_convolution_2d.py

@@ -0,0 +1,18 @@
+import sys, os
+
+sys.path.append(os.path.dirname(os.getcwd()))
+
+from tensorflow import keras
+import numpy as np
+from pyradox import modules
+
+
+def test():
+    inputs = keras.Input(shape=(28, 28, 1))
+    x = modules.Convolution2D(
+        padding="same", activation="relu", batch_normalization=True, dropout=0.2
+    )(inputs)
+    x = keras.layers.GlobalAvgPool2D()(x)
+    outputs = keras.layers.Dense(10, activation="softmax")(x)
+
+    model = keras.models.Model(inputs=inputs, outputs=outputs)

tests/test_dense_net_convolution_block.py

@@ -0,0 +1,20 @@
+import sys, os
+
+sys.path.append(os.path.dirname(os.getcwd()))
+
+from tensorflow import keras
+import numpy as np
+from pyradox import modules
+
+
+def test():
+    inputs = keras.Input(shape=(28, 28, 1))
+    x = keras.layers.Conv2D(
+        3,
+        1,
+    )(inputs)
+    x = modules.DenseNetConvolutionBlock(growth_rate=8, use_bias=True)(x)
+    x = keras.layers.GlobalAvgPool2D()(x)
+    outputs = keras.layers.Dense(10, activation="softmax")(x)
+
+    model = keras.models.Model(inputs=inputs, outputs=outputs)

tests/test_dense_net_transition_block.py

@@ -0,0 +1,21 @@
+import sys, os
+
+sys.path.append(os.path.dirname(os.getcwd()))
+
+from tensorflow import keras
+import numpy as np
+from pyradox import modules
+
+
+def test():
+    inputs = keras.Input(shape=(28, 28, 1))
+    x = keras.layers.Conv2D(
+        3,
+        1,
+    )(inputs)
+    x = modules.DenseNetConvolutionBlock(growth_rate=32, use_bias=True)(x)
+    x = modules.DenseNetTransitionBlock(reduction=0.125)(x)
+    x = keras.layers.GlobalAvgPool2D()(x)
+    outputs = keras.layers.Dense(10, activation="softmax")(x)
+
+    model = keras.models.Model(inputs=inputs, outputs=outputs)

tests/test_dense_skip_connection.py

@@ -0,0 +1,17 @@
+import sys, os
+
+sys.path.append(os.path.dirname(os.getcwd()))
+
+from tensorflow import keras
+import numpy as np
+from pyradox import modules
+
+
+def test():
+    inputs = keras.Input(shape=(13,))
+    x = modules.DenseSkipConnection(
+        32, activation="relu", batch_normalization=True, dropout=0.2
+    )(inputs)
+    outputs = keras.layers.Dense(1)(x)
+
+    model = keras.models.Model(inputs=inputs, outputs=outputs)

tests/test_densely_connected.py

@@ -0,0 +1,20 @@
+import sys, os
+
+sys.path.append(os.path.dirname(os.getcwd()))
+
+from tensorflow import keras
+import numpy as np
+from pyradox import modules
+
+
+def test():
+    inputs = keras.Input(shape=(13,))
+    x = modules.DenselyConnected(
+        32, activation="relu", batch_normalization=True, dropout=0.2
+    )(inputs)
+    x = modules.DenselyConnected(
+        8, activation="relu", batch_normalization=True, dropout=0.2
+    )(x)
+    outputs = keras.layers.Dense(1)(x)
+
+    model = keras.models.Model(inputs=inputs, outputs=outputs)

tests/test_densely_connected_convolutional_network121.py

@@ -0,0 +1,22 @@
+import sys, os
+
+sys.path.append(os.path.dirname(os.getcwd()))
+
+from tensorflow import keras
+import numpy as np
+from pyradox import convnets
+
+
+def test():
+    inputs = keras.Input(shape=(28, 28, 1))
+    x = keras.layers.ZeroPadding2D(32)(
+        inputs
+    )  # padding to increase dimenstions to 92x92
+    x = keras.layers.Conv2D(3, (1, 1), padding="same")(
+        x
+    )  # increasing the number of channels to 3
+    x = convnets.DenselyConnectedConvolutionalNetwork121()(x)
+    x = keras.layers.GlobalAvgPool2D()(x)
+    outputs = keras.layers.Dense(10, activation="softmax")(x)
+
+    model = keras.models.Model(inputs=inputs, outputs=outputs)

tests/test_densely_connected_convolutional_network169.py

@@ -0,0 +1,22 @@
+import sys, os
+
+sys.path.append(os.path.dirname(os.getcwd()))
+
+from tensorflow import keras
+import numpy as np
+from pyradox import convnets
+
+
+def test():
+    inputs = keras.Input(shape=(28, 28, 1))
+    x = keras.layers.ZeroPadding2D(32)(
+        inputs
+    )  # padding to increase dimenstions to 92x92
+    x = keras.layers.Conv2D(3, (1, 1), padding="same")(
+        x
+    )  # increasing the number of channels to 3
+    x = convnets.DenselyConnectedConvolutionalNetwork169()(x)
+    x = keras.layers.GlobalAvgPool2D()(x)
+    outputs = keras.layers.Dense(10, activation="softmax")(x)
+
+    model = keras.models.Model(inputs=inputs, outputs=outputs)

tests/test_densely_connected_convolutional_network201.py

@@ -0,0 +1,22 @@
+import sys, os
+
+sys.path.append(os.path.dirname(os.getcwd()))
+
+from tensorflow import keras
+import numpy as np
+from pyradox import convnets
+
+
+def test():
+    inputs = keras.Input(shape=(28, 28, 1))
+    x = keras.layers.ZeroPadding2D(32)(
+        inputs
+    )  # padding to increase dimenstions to 92x92
+    x = keras.layers.Conv2D(3, (1, 1), padding="same")(
+        x
+    )  # increasing the number of channels to 3
+    x = convnets.DenselyConnectedConvolutionalNetwork201()(x)
+    x = keras.layers.GlobalAvgPool2D()(x)
+    outputs = keras.layers.Dense(10, activation="softmax")(x)
+
+    model = keras.models.Model(inputs=inputs, outputs=outputs)

tests/test_densely_connected_network.py

@@ -0,0 +1,17 @@
+import sys, os
+
+sys.path.append(os.path.dirname(os.getcwd()))
+
+from tensorflow import keras
+import numpy as np
+from pyradox import densenets
+
+
+def test():
+    inputs = keras.Input(shape=(13,))
+    x = densenets.DenselyConnectedNetwork(
+        [32, 8], batch_normalization=True, dropout=0.2
+    )(inputs)
+    outputs = keras.layers.Dense(1)(x)
+
+    model = keras.models.Model(inputs=inputs, outputs=outputs)

tests/test_densely_connected_resnet.py

@@ -0,0 +1,17 @@
+import sys, os
+
+sys.path.append(os.path.dirname(os.getcwd()))
+
+from tensorflow import keras
+import numpy as np
+from pyradox import densenets
+
+
+def test():
+    inputs = keras.Input(shape=(13,))
+    x = densenets.DenselyConnectedResnet(
+        [32, 8], batch_normalization=True, dropout=0.2
+    )(inputs)
+    outputs = keras.layers.Dense(1)(x)
+
+    model = keras.models.Model(inputs=inputs, outputs=outputs)

tests/test_efficient_net.py

@@ -0,0 +1,22 @@
+import sys, os
+
+sys.path.append(os.path.dirname(os.getcwd()))
+
+from tensorflow import keras
+import numpy as np
+from pyradox import convnets
+
+
+def test():
+    inputs = keras.Input(shape=(28, 28, 1))
+    x = keras.layers.ZeroPadding2D(2)(
+        inputs
+    )  # padding to increase dimenstions to 32x32
+    x = keras.layers.Conv2D(3, 1, padding="same")(
+        x
+    )  # increasing the number of channels to 3
+    x = convnets.EfficientNet(1, 1, 32)(x)
+    x = keras.layers.GlobalAvgPool2D()(x)
+    outputs = keras.layers.Dense(10, activation="softmax")(x)
+
+    model = keras.models.Model(inputs=inputs, outputs=outputs)

tests/test_efficient_net_b0.py

@@ -0,0 +1,22 @@
+import sys, os
+
+sys.path.append(os.path.dirname(os.getcwd()))
+
+from tensorflow import keras
+import numpy as np
+from pyradox import convnets
+
+
+def test():
+    inputs = keras.Input(shape=(28, 28, 1))
+    x = keras.layers.ZeroPadding2D(2)(
+        inputs
+    )  # padding to increase dimenstions to 32x32
+    x = keras.layers.Conv2D(3, 1, padding="same")(
+        x
+    )  # increasing the number of channels to 3
+    x = convnets.EfficientNetB0()(x)
+    x = keras.layers.GlobalAvgPool2D()(x)
+    outputs = keras.layers.Dense(10, activation="softmax")(x)
+
+    model = keras.models.Model(inputs=inputs, outputs=outputs)

tests/test_efficient_net_b1.py

@@ -0,0 +1,22 @@
+import sys, os
+
+sys.path.append(os.path.dirname(os.getcwd()))
+
+from tensorflow import keras
+import numpy as np
+from pyradox import convnets
+
+
+def test():
+    inputs = keras.Input(shape=(28, 28, 1))
+    x = keras.layers.ZeroPadding2D(2)(
+        inputs
+    )  # padding to increase dimenstions to 32x32
+    x = keras.layers.Conv2D(3, 1, padding="same")(
+        x
+    )  # increasing the number of channels to 3
+    x = convnets.EfficientNetB1()(x)
+    x = keras.layers.GlobalAvgPool2D()(x)
+    outputs = keras.layers.Dense(10, activation="softmax")(x)
+
+    model = keras.models.Model(inputs=inputs, outputs=outputs)

tests/test_efficient_net_b2.py

@@ -0,0 +1,22 @@
+import sys, os
+
+sys.path.append(os.path.dirname(os.getcwd()))
+
+from tensorflow import keras
+import numpy as np
+from pyradox import convnets
+
+
+def test():
+    inputs = keras.Input(shape=(28, 28, 1))
+    x = keras.layers.ZeroPadding2D(2)(
+        inputs
+    )  # padding to increase dimenstions to 32x32
+    x = keras.layers.Conv2D(3, 1, padding="same")(
+        x
+    )  # increasing the number of channels to 3
+    x = convnets.EfficientNetB2()(x)
+    x = keras.layers.GlobalAvgPool2D()(x)
+    outputs = keras.layers.Dense(10, activation="softmax")(x)
+
+    model = keras.models.Model(inputs=inputs, outputs=outputs)

tests/test_efficient_net_b3.py

@@ -0,0 +1,22 @@
+import sys, os
+
+sys.path.append(os.path.dirname(os.getcwd()))
+
+from tensorflow import keras
+import numpy as np
+from pyradox import convnets
+
+
+def test():
+    inputs = keras.Input(shape=(28, 28, 1))
+    x = keras.layers.ZeroPadding2D(2)(
+        inputs
+    )  # padding to increase dimenstions to 32x32
+    x = keras.layers.Conv2D(3, 1, padding="same")(
+        x
+    )  # increasing the number of channels to 3
+    x = convnets.EfficientNetB3()(x)
+    x = keras.layers.GlobalAvgPool2D()(x)
+    outputs = keras.layers.Dense(10, activation="softmax")(x)
+
+    model = keras.models.Model(inputs=inputs, outputs=outputs)

tests/test_efficient_net_b4.py

@@ -0,0 +1,22 @@
+import sys, os
+
+sys.path.append(os.path.dirname(os.getcwd()))
+
+from tensorflow import keras
+import numpy as np
+from pyradox import convnets
+
+
+def test():
+    inputs = keras.Input(shape=(28, 28, 1))
+    x = keras.layers.ZeroPadding2D(2)(
+        inputs
+    )  # padding to increase dimenstions to 32x32
+    x = keras.layers.Conv2D(3, 1, padding="same")(
+        x
+    )  # increasing the number of channels to 3
+    x = convnets.EfficientNetB4()(x)
+    x = keras.layers.GlobalAvgPool2D()(x)
+    outputs = keras.layers.Dense(10, activation="softmax")(x)
+
+    model = keras.models.Model(inputs=inputs, outputs=outputs)

tests/test_efficient_net_b5.py

@@ -0,0 +1,18 @@
+import sys, os
+
+sys.path.append(os.path.dirname(os.getcwd()))
+
+from tensorflow import keras
+import numpy as np
+from pyradox import convnets
+
+def test():
+    inputs = keras.Input(shape=(28, 28, 1))
+    x = keras.layers.ZeroPadding2D(2)(inputs)         # padding to increase dimenstions to 32x32
+    x = keras.layers.Conv2D(3, 1, padding='same')(x)  # increasing the number of channels to 3
+    x = convnets.EfficientNetB5()(x)
+    x = keras.layers.GlobalAvgPool2D()(x)
+    outputs = keras.layers.Dense(10, activation="softmax")(x)
+
+    model = keras.models.Model(inputs=inputs, outputs=outputs) 
+