ml-clusters/utils.py at main · Sulla2012/ml-clusters · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
import numpy as np
from numba import jit
import tensorflow as tf
from tensorflow.keras import models, layers, regularizers

@jit(nopython=True)
def normalize_map(imap):
    temp_map = np.zeros(imap.shape)
    for j in range(imap.shape[-1]):
        temp = (imap[...,j]-np.mean(imap[...,j]))/np.std(imap[...,j])
        temp_map[...,j] = temp

    return temp_map

#A helper function that stores a bunch of models I've been playing around with. Notable Kosiba is a model following
#Kosiba 2020 (https://arxiv.org/abs/2006.05998) who had good success training it on x-ray+ optical data
def make_model(model_type = 'simple_cnn', input_shape = (342,342,5), degree = 5):

    if model_type == 'simple_cnn':
        model = models.Sequential()
        model.add(layers.Conv2D(32, (3, 3), activation='relu', input_shape=input_shape))
        model.add(layers.MaxPooling2D((2, 2)))
        model.add(layers.Conv2D(64, (3, 3), activation='relu'))
        model.add(layers.MaxPooling2D((2, 2)))
        model.add(layers.Conv2D(64, (3, 3), activation='relu'))
        model.add(layers.Flatten())
        model.add(layers.Dense(64, activation='relu'))
        model.add(layers.Dense(2))
    elif model_type == 'resnet':
        model = tf.keras.applications.ResNet50(include_top=True,
        weights= None,
        input_shape = input_shape,
        classes = 2)
    elif model_type == 'pretrained':
        model = tf.keras.applications.ResNet50(include_top=False,
        weights= 'imagenet',
        input_shape = input_shape_rgb,
        classes = 2)

    elif model_type == 'inception_v2':
        base_model = tf.keras.applications.InceptionResNetV2(include_top = False,
                input_shape = input_shape,
                weights= 'imagenet'
                )
        base_model.trainable = False


    elif model_type == 'hyper':
        def train_test_model(hparams):
            model = models.Sequential()
            model.add(layers.Conv2D(32, (3, 3), activation='relu', input_shape=input_shape))
            model.add(layers.MaxPooling2D((2, 2)))
            model.add(layers.Conv2D(64, (3, 3), activation='relu'))
            model.add(layers.MaxPooling2D((2, 2)))
            model.add(layers.Dropout(hparams[HP_DROPOUT]))
            model.add(layers.Conv2D(64, (3, 3), activation='relu'))
            model.add(layers.Flatten())
            model.add(layers.Dense(hparams[HP_NUM_UNITS], activation='relu'))
            model.add(layers.Dense(2))
            model.compile(
              optimizer=hparams[HP_OPTIMIZER],
              loss='sparse_categorical_crossentropy',
              metrics=['accuracy'],
          )

            model.fit(train_dataset, epochs=10)
            _, accuracy = model.evaluate(test_dataset)
            return accuracy


    elif model_type == 'test':
        model = models.Sequential([
        layers.Conv2D(16, 3, padding='same', activation='relu',
               input_shape=input_shape),
        layers.MaxPooling2D(),
        layers.Dropout(0.2),
        layers.Conv2D(32, 3, padding='same', activation='relu'),
        layers.MaxPooling2D(),
        layers.Conv2D(64, 3, padding='same', activation='relu'),
        layers.MaxPooling2D(),
        layers.Dropout(0.2),
        layers.Flatten(),
        layers.Dense(512, activation='relu'),
        layers.Dense(2)])

    elif model_type == 'deep_few_filters':
        model = models.Sequential()
        #model.add(layers.experimental.preprocessing.Rescaling(1./all_max, input_shape=input_shape))
        model.add(layers.Conv2D(2, (3, 3), activation='relu', input_shape=input_shape))
        model.add(layers.MaxPooling2D((2, 2)))
        j = 0
        for j in range(degree):
            model.add(layers.Conv2D(4, (3, 3), activation='relu'))
            model.add(layers.ZeroPadding2D((1,1)))
            #model.add(layers.MaxPooling2D((2, 2)))
            j +=1
        model.add(layers.Conv2D(8, (3, 3), activation='relu'))
        model.add(layers.Flatten())
        model.add(layers.Dense(16, activation='relu'))
        model.add(layers.Dense(2))

    elif model_type == 'dropout':
        model = models.Sequential()
        model.add(layers.Conv2D(32, (3, 3), kernel_regularizer=regularizers.l2(0.0001), activation='relu', input_shape=input_shape))
        layers.Dropout(0.5),
        model.add(layers.AveragePooling2D((2, 2)))
        model.add(layers.Conv2D(64, (3, 3), kernel_regularizer=regularizers.l2(0.0001), activation='relu'))
        layers.Dropout(0.5),
        model.add(layers.AveragePooling2D((2, 2)))
        model.add(layers.Conv2D(64, (3, 3), kernel_regularizer=regularizers.l2(0.0001), activation='relu'))
        layers.Dropout(0.5),
        model.add(layers.Flatten())
        model.add(layers.Dense(64, activation='relu'))
        model.add(layers.Dense(2))


    elif model_type == 'kosiba':
        model = models.Sequential()
        model.add(layers.Conv2D(64, (3, 3), activation='relu', input_shape=input_shape))
        model.add(layers.MaxPooling2D((2, 2), strides=(2,2)))
        for i in range(degree):
            model.add(layers.Conv2D(32, (3, 3), activation='relu'))
            model.add(layers.MaxPooling2D((2, 2), strides=(2,2)))
        model.add(layers.Flatten())
        model.add(layers.Dense(256, activation='relu', kernel_regularizer=regularizers.l1(0.0001)))
        layers.Dropout(0.65)

        model.add(layers.Dense(2))
    model.summary()
    if model_type == 'kosiba':
        lr_schedule = tf.keras.optimizers.schedules.InverseTimeDecay(
              0.0001,
              decay_steps=348/10*1000,
              decay_rate=1e-6,
              staircase=False)
        opt = tf.keras.optimizers.SGD(
                learning_rate=0.01, momentum=0.9, nesterov = True, name='SGD')
        model.compile(optimizer=opt,
                  loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True),
                  metrics=['accuracy'])
    else:
        model.compile(optimizer='Adam',
              loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True),
              metrics=['accuracy'])
    return model

class feature_model:

    def __init__(self, train, comp, w_y = False):
        self.w_y = w_y
        self.train = clean_inputs(train)
        self.comp = clean_inputs(comp)

    def clean_inputs(data_tensor):
        flags = []
        for i in range(data_tensor.shape[0]):
            if np.any(np.isnan(data_tensor[i,...])):
                flags.append(i)
        data_tensor = np.delete(data_tensor, flags, axis = 0)