CoCalc Logo Icon
StoreFeaturesDocsShareSupportNewsAboutSign UpSign In
amanchadha

Real-time collaboration for Jupyter Notebooks, Linux Terminals, LaTeX, VS Code, R IDE, and more,
all in one place. Commercial Alternative to JupyterHub.

GitHub Repository: amanchadha/coursera-deep-learning-specialization
 Path: blob/master/C5 - Sequence Models/Week 3/Trigger word detection/test_utils.py
Views: 4819
1
from termcolor import colored

2


3
from tensorflow.keras.layers import Input

4
from tensorflow.keras.layers import Conv2D

5
from tensorflow.keras.layers import MaxPooling2D

6
from tensorflow.keras.layers import Dropout 

7
from tensorflow.keras.layers import Conv2DTranspose

8
from tensorflow.keras.layers import concatenate

9
from tensorflow.keras.layers import ZeroPadding2D

10
from tensorflow.keras.layers import Dense

11
from tensorflow.keras.layers import LSTM

12
from tensorflow.keras.layers import RepeatVector

13
from tensorflow.keras.layers import TimeDistributed

14
from tensorflow.keras.layers import GRU

15
from tensorflow.keras.layers import Conv1D

16


17
# Compare the two inputs

18
def comparator(learner, instructor):

19
    layer = 0

20
    for a, b in zip(learner, instructor):

21
        if tuple(a) != tuple(b):

22
            print(colored("Test failed", attrs=['bold']),

23
                  f"at layer: {layer}",

24
                  "\n Expected value \n\n", colored(f"{b}", "green"), 

25
                  "\n\n does not match the input value: \n\n", 

26
                  colored(f"{a}", "red"))

27
            raise AssertionError("Error in test") 

28
        layer += 1

29
    print(colored("All tests passed!", "green"))

30


31
# extracts the description of a given model

32
def summary(model):

33
    model.compile(optimizer='adam',

34
                  loss='categorical_crossentropy',

35
                  metrics=['accuracy'])

36
    result = []

37
    for layer in model.layers:

38
        descriptors = [layer.__class__.__name__, layer.output_shape, layer.count_params()]

39
        if (type(layer) == Conv1D):

40
            descriptors.append(layer.padding)

41
            descriptors.append(layer.activation.__name__)

42
            descriptors.append(layer.strides)

43
            descriptors.append(layer.kernel_size)

44
            descriptors.append(layer.kernel_initializer.__class__.__name__)

45
           

46
        if (type(layer) == Conv2D):

47
            descriptors.append(layer.padding)

48
            descriptors.append(layer.activation.__name__)

49
            descriptors.append(layer.kernel_initializer.__class__.__name__)

50
            

51
        if (type(layer) == MaxPooling2D):

52
            descriptors.append(layer.pool_size)

53
            descriptors.append(layer.strides)

54
            descriptors.append(layer.padding)

55
            

56
        if (type(layer) == Dropout):

57
            descriptors.append(layer.rate)

58
            

59
        if (type(layer) == ZeroPadding2D):

60
            descriptors.append(layer.padding)

61
            

62
        if (type(layer) == Dense):

63
            descriptors.append(layer.activation.__name__)

64
            

65
        if (type(layer) == LSTM):

66
            descriptors.append(layer.input_shape)

67
            descriptors.append(layer.activation.__name__)

68
            

69
        if (type(layer) == RepeatVector):

70
            descriptors.append(layer.n)

71
            

72
        if (type(layer) == TimeDistributed):

73
            descriptors.append(layer.layer.activation.__name__)  

74
            

75
        if (type(layer) == GRU):

76
            descriptors.append(layer.return_sequences)    

77
            

78
        result.append(descriptors)

79
    return result

80