Code DD

3/5/24, 1:01 AM Untitled
In [75]: import tensorflow as tf

import matplotlib.pyplot as plt
import numpy as np
tf.__version__
Out[75]: '2.10.1'
In [76]: base_model = tf.keras.applications.InceptionV3(include_top = False, weights =

'imagenet')
In [77]: base_model.summary()
localhost:8889/nbconvert/html/Dropbox/Code Practice/CV masterclass/Untitled.ipynb?download=false 1/47

Model: "inception_v3"
____________________________________________________________________________________
______________
Layer (type) Output Shape Param # Connected to
====================================================================================
==============
input_3 (InputLayer) [(None, None, None, 0 []
3)]
conv2d_188 (Conv2D) (None, None, None, 864 ['input_3[0][0]']

32)
batch_normalization_188 (Batch (None, None, None, 96 ['conv2d_188[0]

[0]']
Normalization) 32)
activation_188 (Activation) (None, None, None, 0 ['batch_normalizati

on_188[0][0]']
32)
conv2d_189 (Conv2D) (None, None, None, 9216 ['activation_188[0]

[0]']
32)

[0]']
Normalization) 32)

on_189[0][0]']
32)

[0]']
64)

[0]']
Normalization) 64)

on_190[0][0]']
64)
max_pooling2d_8 (MaxPooling2D) (None, None, None, 0 ['activation_190[0]

[0]']
64)
conv2d_191 (Conv2D) (None, None, None, 5120 ['max_pooling2d_8

[0][0]']
80)

[0]']
Normalization) 80)


on_191[0][0]']
80)

[0]']
192)

[0]']
Normalization) 192)

on_192[0][0]']
192)
max_pooling2d_9 (MaxPooling2D) (None, None, None, 0 ['activation_192[0]

[0]']
192)

[0][0]']
64)

[0]']
Normalization) 64)

on_196[0][0]']
64)

[0][0]']
48)

[0]']
96)

[0]']
Normalization) 48)

[0]']
Normalization) 96)

on_194[0][0]']
48)

on_197[0][0]']
96)

average_pooling2d_18 (AverageP (None, None, None, 0 ['max_pooling2d_9

[0][0]']
ooling2D) 192)

[0][0]']
64)

[0]']
64)

[0]']
96)
conv2d_199 (Conv2D) (None, None, None, 6144 ['average_pooling2d

_18[0][0]']
32)

[0]']
Normalization) 64)

[0]']
Normalization) 64)

[0]']
Normalization) 96)

[0]']
Normalization) 32)

on_193[0][0]']
64)

on_195[0][0]']
64)

on_198[0][0]']
96)

on_199[0][0]']
32)
mixed0 (Concatenate) (None, None, None, 0 ['activation_193[0]

[0]',
256) 'activation_195[0]
[0]',

'activation_198[0]
[0]',
'activation_199[0]
[0]']
conv2d_203 (Conv2D) (None, None, None, 16384 ['mixed0[0][0]']

64)

[0]']
Normalization) 64)

on_203[0][0]']
64)

48)

[0]']
96)

[0]']
Normalization) 48)

[0]']
Normalization) 96)

on_201[0][0]']
48)

on_204[0][0]']
96)
average_pooling2d_19 (AverageP (None, None, None, 0 ['mixed0[0][0]']

ooling2D) 256)

64)

[0]']
64)

[0]']
96)

_19[0][0]']
64)


[0]']
Normalization) 64)

[0]']
Normalization) 64)

[0]']
Normalization) 96)

[0]']
Normalization) 64)

on_200[0][0]']
64)

on_202[0][0]']
64)

on_205[0][0]']
96)

on_206[0][0]']
64)

[0]',
[0]',
'activation_205[0]
[0]',
'activation_206[0]
[0]']

64)

[0]']
Normalization) 64)

on_210[0][0]']
64)

48)


[0]']
96)

[0]']
Normalization) 48)

[0]']
Normalization) 96)

on_208[0][0]']
48)

on_211[0][0]']
96)

ooling2D) 288)

64)

[0]']
64)

[0]']
96)

_20[0][0]']
64)

[0]']
Normalization) 64)

[0]']
Normalization) 64)

[0]']
Normalization) 96)

[0]']
Normalization) 64)

on_207[0][0]']

64)

on_209[0][0]']
64)

on_212[0][0]']
96)

on_213[0][0]']
64)

[0]',
[0]',
'activation_212[0]
[0]',
'activation_213[0]
[0]']

64)

[0]']
Normalization) 64)

on_215[0][0]']
64)

[0]']
96)

[0]']
Normalization) 96)

on_216[0][0]']
96)

384)

[0]']
96)

[0]']
Normalization) 384)


[0]']
Normalization) 96)

on_214[0][0]']
384)

on_217[0][0]']
96)
max_pooling2d_10 (MaxPooling2D (None, None, None, 0 ['mixed2[0][0]']

) 288)

[0]',
[0]',
'max_pooling2d_10
[0][0]']

128)

[0]']
Normalization) 128)

on_222[0][0]']
128)

[0]']
128)

[0]']
Normalization) 128)

on_223[0][0]']
128)

128)

[0]']
128)

[0]']
Normalization) 128)


[0]']
Normalization) 128)

on_219[0][0]']
128)

on_224[0][0]']
128)

[0]']
128)

[0]']
128)

[0]']
Normalization) 128)

[0]']
Normalization) 128)

on_220[0][0]']
128)

on_225[0][0]']
128)

ooling2D) 768)

192)

[0]']
192)

[0]']
192)

_21[0][0]']
192)

[0]']
Normalization) 192)

[0]']
Normalization) 192)

[0]']
Normalization) 192)

[0]']
Normalization) 192)

on_218[0][0]']
192)

on_221[0][0]']
192)

on_226[0][0]']
192)

on_227[0][0]']
192)

[0]',
[0]',
'activation_226[0]
[0]',
'activation_227[0]
[0]']

160)

[0]']
Normalization) 160)

on_232[0][0]']
160)

[0]']
160)

[0]']
Normalization) 160)

on_233[0][0]']
160)

160)

[0]']
160)

[0]']
Normalization) 160)

[0]']
Normalization) 160)

on_229[0][0]']
160)

on_234[0][0]']
160)

[0]']
160)

[0]']
160)

[0]']
Normalization) 160)

[0]']
Normalization) 160)

on_230[0][0]']
160)

on_235[0][0]']
160)

ooling2D) 768)


192)

[0]']
192)

[0]']
192)

_22[0][0]']
192)

[0]']
Normalization) 192)

[0]']
Normalization) 192)

[0]']
Normalization) 192)

[0]']
Normalization) 192)

on_228[0][0]']
192)

on_231[0][0]']
192)

on_236[0][0]']
192)

on_237[0][0]']
192)

[0]',
[0]',
'activation_236[0]
[0]',
'activation_237[0]
[0]']


160)

[0]']
Normalization) 160)

on_242[0][0]']
160)

[0]']
160)

[0]']
Normalization) 160)

on_243[0][0]']
160)

160)

[0]']
160)

[0]']
Normalization) 160)

[0]']
Normalization) 160)

on_239[0][0]']
160)

on_244[0][0]']
160)

[0]']
160)

[0]']
160)

[0]']
Normalization) 160)

[0]']
Normalization) 160)

on_240[0][0]']
160)

on_245[0][0]']
160)

ooling2D) 768)

192)

[0]']
192)

[0]']
192)

_23[0][0]']
192)

[0]']
Normalization) 192)

[0]']
Normalization) 192)

[0]']
Normalization) 192)

[0]']
Normalization) 192)

on_238[0][0]']
192)

on_241[0][0]']
192)


on_246[0][0]']
192)

on_247[0][0]']
192)

[0]',
[0]',
'activation_246[0]
[0]',
'activation_247[0]
[0]']

192)

[0]']
Normalization) 192)

on_252[0][0]']
192)

[0]']
192)

[0]']
Normalization) 192)

on_253[0][0]']
192)

192)

[0]']
192)

[0]']
Normalization) 192)

[0]']
Normalization) 192)


on_249[0][0]']
192)

on_254[0][0]']
192)

[0]']
192)

[0]']
192)

[0]']
Normalization) 192)

[0]']
Normalization) 192)

on_250[0][0]']
192)

on_255[0][0]']
192)

ooling2D) 768)

192)

[0]']
192)

[0]']
192)

_24[0][0]']
192)

[0]']
Normalization) 192)

[0]']

Normalization) 192)

[0]']
Normalization) 192)

[0]']
Normalization) 192)

on_248[0][0]']
192)

on_251[0][0]']
192)

on_256[0][0]']
192)

on_257[0][0]']
192)

[0]',
[0]',
'activation_256[0]
[0]',
'activation_257[0]
[0]']

192)

[0]']
Normalization) 192)

on_260[0][0]']
192)

[0]']
192)

[0]']
Normalization) 192)

on_261[0][0]']

192)

192)

[0]']
192)

[0]']
Normalization) 192)

[0]']
Normalization) 192)

on_258[0][0]']
192)

on_262[0][0]']
192)

[0]']
320)

[0]']
192)

[0]']
Normalization) 320)

[0]']
Normalization) 192)

on_259[0][0]']
320)

on_263[0][0]']
192)
max_pooling2d_11 (MaxPooling2D (None, None, None, 0 ['mixed7[0][0]']

) 768)

[0]',
[0]',

'max_pooling2d_11
[0][0]']

448)

[0]']
Normalization) 448)

on_268[0][0]']
448)

384)

[0]']
384)

[0]']
Normalization) 384)

[0]']
Normalization) 384)

on_265[0][0]']
384)

on_269[0][0]']
384)

[0]']
384)

[0]']
384)

[0]']
384)

[0]']
384)

ooling2D) 1280)


320)

[0]']
Normalization) 384)

[0]']
Normalization) 384)

[0]']
Normalization) 384)

[0]']
Normalization) 384)

_25[0][0]']
192)

[0]']
Normalization) 320)

on_266[0][0]']
384)

on_267[0][0]']
384)

on_270[0][0]']
384)

on_271[0][0]']
384)

[0]']
Normalization) 192)

on_264[0][0]']
320)
mixed9_0 (Concatenate) (None, None, None, 0 ['activation_266[0]

[0]',
[0]']

concatenate_4 (Concatenate) (None, None, None, 0 ['activation_270[0]

[0]',
[0]']

on_272[0][0]']
192)

[0]',
2048) 'mixed9_0[0][0]',
'concatenate_4[0]
[0]',
'activation_272[0]
[0]']

448)

[0]']
Normalization) 448)

on_277[0][0]']
448)

384)

[0]']
384)

[0]']
Normalization) 384)

[0]']
Normalization) 384)

on_274[0][0]']
384)

on_278[0][0]']
384)

[0]']
384)

[0]']
384)

[0]']
384)

[0]']
384)

ooling2D) 2048)

320)

[0]']
Normalization) 384)

[0]']
Normalization) 384)

[0]']
Normalization) 384)

[0]']
Normalization) 384)

_26[0][0]']
192)

[0]']
Normalization) 320)

on_275[0][0]']
384)

on_276[0][0]']
384)

on_279[0][0]']
384)

on_280[0][0]']
384)


[0]']
Normalization) 192)

on_273[0][0]']
320)
mixed9_1 (Concatenate) (None, None, None, 0 ['activation_275[0]

[0]',
[0]']
concatenate_5 (Concatenate) (None, None, None, 0 ['activation_279[0]

[0]',
[0]']

on_281[0][0]']
192)

[0]',
2048) 'mixed9_1[0][0]',
'concatenate_5[0]
[0]',
'activation_281[0]
[0]']
====================================================================================
==============
Total params: 21,802,784
Trainable params: 21,768,352
Non-trainable params: 34,432
____________________________________________________________________________________
______________
In [78]: len(base_model.layers)
Out[78]: 311
In [79]: names = ['mixed3', 'mixed5']
In [80]: base_model.input
Out[80]: <KerasTensor: shape=(None, None, None, 3) dtype=float32 (created by layer 'input_

3')>
In [81]: layers = [base_model.get_layer(name).output for name in names]

layers

Out[81]: [<KerasTensor: shape=(None, None, None, 768) dtype=float32 (created by layer 'mixe
d3')>,
<KerasTensor: shape=(None, None, None, 768) dtype=float32 (created by layer 'mixe
d5')>]
In [82]: deep_dream_model = tf.keras.Model(inputs = base_model.input, outputs = layers )
In [83]: image = tf.keras.preprocessing.image.load_img("G:\Downloads (2024)\Computer

Vision Masterclass-20240304T131702Z-001\Computer Vision
Masterclass\Images\StaryNight.jpg",
target_size = (225,375))
In [84]: plt.imshow(image)
Out[84]: <matplotlib.image.AxesImage at 0x18cac237a00>
In [85]: print(type(image)) #format is in pillow, so we will convert it into numpy format

for our NN
print(image.mode )
<class 'PIL.Image.Image'>
RGB
In [86]: print(image.mode )
print(len(image.mode))
RGB
3
In [87]: list(image.getdata())

Out[87]: [(6, 10, 11),

(30, 33, 66),
(35, 46, 66),
(39, 46, 65),
(46, 49, 68),
(23, 27, 39),
(13, 15, 38),
(57, 61, 86),
(85, 88, 97),
(147, 147, 139),
(138, 139, 123),
(112, 116, 117),
(18, 21, 52),
(23, 28, 60),
(85, 96, 116),
(83, 92, 121),
(148, 152, 155),
(29, 32, 63),
(57, 63, 99),
(12, 20, 39),
(34, 39, 69),
(23, 30, 56),
(4, 12, 48),
(25, 32, 74),
(48, 50, 98),
(134, 133, 173),
(127, 132, 152),
(126, 133, 161),
(138, 151, 168),
(64, 76, 126),
(67, 83, 119),
(61, 75, 112),
(73, 90, 134),
(86, 118, 143),
(55, 81, 80),
(136, 155, 135),
(51, 65, 74),
(152, 169, 195),
(156, 176, 211),
(56, 79, 129),
(94, 116, 165),
(70, 94, 120),
(160, 173, 189),
(142, 156, 185),
(92, 118, 135),
(50, 74, 118),
(107, 132, 162),
(101, 122, 153),
(27, 36, 67),
(34, 44, 79),
(97, 114, 140),
(21, 41, 66),
(51, 60, 101),
(50, 56, 104),
(34, 39, 68),
(22, 24, 45),

(3, 3, 29),
(4, 7, 22),
(28, 37, 34),
(19, 27, 29),
(19, 25, 41),
(43, 49, 75),
(93, 104, 122),
(84, 96, 122),
(33, 39, 71),
(16, 19, 52),
(14, 17, 48),
(18, 25, 54),
(99, 108, 139),
(63, 66, 99),
(46, 53, 79),
(72, 77, 117),
(20, 22, 60),
(78, 81, 90),
(184, 173, 169),
(137, 128, 119),
(102, 93, 96),
(182, 176, 160),
(141, 130, 134),
(136, 134, 111),
(49, 57, 78),
(33, 50, 102),
(43, 58, 89),
(36, 46, 73),
(51, 63, 101),
(79, 99, 126),
(45, 69, 97),
(40, 54, 93),
(67, 78, 123),
(64, 82, 120),
(81, 86, 116),
(75, 83, 130),
(75, 88, 140),
(54, 71, 115),
(54, 70, 122),
(46, 60, 109),
(42, 54, 102),
(39, 50, 95),
(31, 39, 76),
(14, 20, 52),
(25, 30, 60),
(15, 19, 46),
(175, 177, 198),
(99, 101, 124),
(15, 18, 51),
(22, 27, 65),
(46, 53, 105),
(42, 47, 103),
(75, 80, 120),
(93, 102, 131),
(122, 131, 170),
(86, 110, 182),

(50, 75, 129),

(49, 63, 112),
(79, 90, 135),
(51, 69, 135),
(43, 56, 111),
(37, 51, 98),
(84, 98, 143),
(187, 201, 230),
(67, 81, 116),
(31, 46, 101),
(54, 63, 128),
(75, 81, 115),
(39, 56, 74),
(105, 126, 147),
(113, 122, 129),
(138, 146, 149),
(92, 96, 123),
(149, 148, 154),
(190, 186, 177),
(131, 135, 136),
(89, 105, 104),
(71, 86, 129),
(74, 92, 132),
(89, 94, 150),
(64, 62, 67),
(66, 64, 69),
(28, 30, 55),
(77, 80, 133),
(2, 7, 36),
(111, 106, 112),
(181, 174, 148),
(66, 73, 99),
(59, 83, 143),
(56, 69, 114),
(43, 52, 117),
(79, 83, 110),
(162, 151, 167),
(152, 145, 135),
(176, 171, 178),
(99, 99, 111),
(72, 81, 90),
(95, 109, 120),
(77, 89, 115),
(66, 76, 112),
(60, 74, 100),
(70, 84, 119),
(39, 51, 99),
(24, 33, 74),
(40, 47, 76),
(48, 52, 89),
(39, 42, 83),
(34, 39, 71),
(18, 15, 26),
(117, 118, 102),
(167, 167, 155),
(126, 120, 132),

(90, 85, 92),

(101, 99, 84),
(127, 121, 121),
(137, 133, 121),
(155, 154, 149),
(165, 166, 170),
(73, 79, 93),
(78, 91, 99),
(110, 117, 135),
(160, 155, 149),
(179, 166, 131),
(164, 148, 133),
(151, 144, 136),
(108, 104, 103),
(126, 124, 129),
(44, 39, 46),
(186, 182, 170),
(56, 62, 78),
(117, 119, 118),
(51, 51, 89),
(75, 78, 95),
(145, 128, 120),
(164, 144, 117),
(143, 132, 112),
(165, 164, 146),
(125, 127, 126),
(110, 121, 127),
(143, 152, 167),
(124, 125, 127),
(80, 92, 140),
(74, 95, 152),
(77, 95, 133),
(103, 109, 143),
(124, 128, 137),
(134, 133, 115),
(126, 137, 129),
(116, 140, 152),
(114, 125, 131),
(106, 104, 128),
(196, 190, 176),
(128, 123, 127),
(141, 142, 146),
(113, 115, 127),
(87, 89, 127),
(83, 89, 113),
(95, 102, 121),
(89, 95, 127),
(82, 88, 120),
(74, 81, 107),
(50, 54, 92),
(41, 47, 83),
(46, 55, 84),
(34, 44, 69),
(32, 39, 67),
(61, 62, 83),
(85, 85, 93),

(131, 129, 130),

(130, 125, 121),
(128, 122, 110),
(166, 157, 152),
(120, 113, 103),
(143, 136, 118),
(181, 174, 158),
(143, 134, 117),
(153, 146, 130),
(101, 98, 93),
(77, 82, 76),
(46, 49, 68),
(30, 31, 52),
(136, 137, 132),
(105, 98, 79),
(81, 77, 66),
(162, 159, 152),
(137, 134, 127),
(184, 180, 177),
(130, 130, 128),
(141, 141, 139),
(124, 124, 116),
(133, 130, 115),
(170, 163, 147),
(130, 128, 115),
(121, 119, 120),
(142, 140, 127),
(128, 129, 121),
(78, 80, 103),
(65, 69, 104),
(46, 53, 99),
(50, 58, 95),
(79, 86, 128),
(78, 83, 123),
(59, 65, 91),
(76, 83, 102),
(106, 107, 138),
(151, 153, 168),
(74, 75, 77),
(95, 95, 103),
(99, 105, 117),
(93, 98, 118),
(101, 105, 117),
(105, 110, 113),
(133, 137, 162),
(55, 61, 113),
(34, 47, 99),
(78, 88, 141),
(57, 65, 128),
(40, 48, 120),
(56, 66, 135),
(64, 78, 141),
(78, 90, 140),
(87, 99, 147),
(73, 84, 130),
(90, 101, 129),

(73, 92, 134),

(122, 125, 134),
(169, 165, 140),
(179, 176, 159),
(116, 115, 113),
(101, 105, 114),
(78, 83, 112),
(97, 102, 124),
(78, 86, 109),
(108, 116, 152),
(100, 113, 132),
(102, 109, 128),
(115, 118, 133),
(117, 114, 121),
(144, 138, 138),
(174, 163, 159),
(143, 144, 149),
(79, 78, 96),
(90, 89, 105),
(123, 122, 117),
(127, 122, 128),
(138, 141, 134),
(108, 102, 106),
(114, 109, 106),
(79, 85, 117),
(84, 93, 108),
(109, 108, 114),
(143, 142, 138),
(124, 117, 111),
(151, 142, 135),
(130, 129, 124),
(112, 115, 124),
(106, 118, 142),
(114, 118, 147),
(107, 116, 147),
(95, 112, 140),
(123, 135, 151),
(161, 162, 154),
(125, 131, 153),
(81, 109, 149),
(65, 94, 150),
(94, 116, 166),
(159, 181, 204),
(100, 120, 155),
(153, 171, 193),
(140, 153, 161),
(171, 175, 178),
(218, 220, 199),
(121, 120, 126),
(91, 97, 123),
(98, 114, 137),
(73, 88, 129),
(89, 106, 134),
(76, 96, 131),
(116, 142, 167),
(96, 122, 155),

(120, 148, 198),

(110, 141, 187),
(119, 149, 185),
(116, 144, 181),
(48, 71, 113),
(55, 77, 127),
(41, 60, 118),
(51, 69, 131),
(44, 59, 116),
(29, 45, 97),
(19, 33, 82),
(3, 15, 65),
(26, 38, 86),
(34, 45, 107),
(33, 44, 100),
(80, 92, 140),
(79, 93, 132),
(54, 67, 99),
(46, 60, 99),
(38, 44, 80),
(71, 81, 130),
(112, 143, 189),
(103, 138, 166),
(126, 154, 193),
(77, 102, 143),
(43, 62, 118),
(42, 58, 117),
(61, 73, 131),
(75, 83, 132),
(44, 44, 70),
(107, 98, 99),
(187, 174, 165),
(96, 86, 94),
(28, 25, 52),
(166, 151, 122),
(217, 197, 160),
(174, 154, 130),
(159, 145, 110),
(191, 177, 138),
(13, 14, 34),
(30, 32, 73),
(25, 32, 61),
(21, 26, 55),
(21, 23, 46),
(5, 6, 37),
(19, 19, 57),
(45, 49, 78),
(22, 26, 38),
(19, 22, 27),
(13, 14, 19),
(108, 105, 124),
(15, 12, 39),
(7, 7, 31),
(44, 49, 68),
(74, 79, 111),
(108, 112, 89),

(126, 130, 141),

(23, 32, 63),
(43, 58, 97),
(28, 44, 96),
(54, 58, 95),
(73, 79, 111),
(49, 55, 103),
(96, 102, 138),
(114, 124, 123),
(165, 166, 170),
(146, 150, 153),
(137, 144, 170),
(82, 95, 137),
(44, 58, 103),
(50, 70, 95),
(143, 171, 149),
(169, 189, 138),
(152, 160, 109),
(142, 151, 134),
(119, 141, 154),
(161, 173, 197),
(159, 177, 197),
(87, 109, 150),
(93, 115, 162),
(113, 134, 179),
(105, 125, 149),
(83, 107, 94),
(113, 129, 152),
(114, 129, 134),
(34, 41, 69),
(75, 81, 117),
(36, 42, 74),
(22, 19, 50),
(43, 43, 93),
(81, 97, 122),
(108, 132, 134),
(54, 72, 94),
(0, 10, 26),
(13, 21, 40),
(25, 28, 47),
(20, 25, 31),
(9, 8, 16),
(22, 19, 40),
(16, 14, 36),
(32, 35, 50),
(17, 27, 37),
(55, 72, 115),
(77, 94, 138),
(27, 37, 73),
(6, 12, 48),
(84, 89, 131),
(74, 79, 121),
(100, 111, 131),
(125, 138, 157),
(47, 56, 95),
(54, 60, 112),

(48, 52, 97),

(66, 79, 123),
(50, 61, 107),
(42, 52, 114),
(47, 59, 109),
(49, 64, 121),
(38, 57, 99),
(64, 90, 127),
(77, 107, 133),
(106, 132, 157),
(47, 73, 108),
(95, 127, 165),
(57, 90, 125),
(94, 116, 163),
(4, 14, 39),
(28, 31, 40),
(21, 23, 46),
(101, 132, 153),
(15, 30, 51),
(20, 26, 38),
(22, 28, 42),
(21, 40, 73),
(30, 42, 82),
(23, 34, 79),
(79, 89, 138),
(44, 54, 105),
(32, 42, 93),
(79, 97, 147),
(58, 75, 129),
(55, 69, 130),
(47, 61, 126),
(30, 45, 112),
(64, 80, 142),
(49, 61, 121),
(35, 45, 104),
(5, 21, 47),
(40, 55, 86),
(112, 129, 175),
(46, 71, 151),
(85, 109, 179),
(27, 44, 100),
(78, 92, 141),
(16, 34, 98),
(76, 98, 155),
(18, 42, 88),
(83, 108, 128),
(62, 83, 104),
(52, 70, 154),
(15, 47, 134),
(77, 110, 151),
(60, 85, 126),
(30, 53, 105),
(43, 61, 125),
(110, 127, 143),
(86, 100, 126),
(134, 159, 120),

(146, 172, 83),

(160, 180, 95),
(72, 83, 113),
(122, 132, 157),
(108, 134, 125),
(78, 109, 140),
(88, 125, 170),
(65, 104, 159),
(76, 99, 201),
(86, 113, 160),
(30, 50, 113),
(13, 29, 62),
(79, 83, 154),
(46, 53, 97),
(92, 100, 149),
(46, 65, 141),
(11, 43, 116),
(69, 87, 163),
(0, 3, 53),
(14, 31, 87),
(33, 48, 105),
(60, 73, 141),
(67, 84, 153),
(58, 79, 136),
(57, 86, 146),
(61, 90, 158),
(76, 100, 170),
(75, 100, 154),
(69, 88, 154),
(66, 83, 161),
(56, 74, 150),
(61, 73, 147),
(10, 29, 87),
(36, 59, 113),
(23, 38, 107),
(42, 64, 124),
(37, 58, 115),
(32, 53, 110),
(27, 49, 90),
(56, 80, 106),
(60, 81, 110),
(82, 98, 124),
(56, 63, 89),
(178, 183, 176),
(132, 133, 125),
(129, 143, 170),
(89, 109, 142),
(22, 37, 94),
(63, 76, 111),
(66, 83, 103),
(42, 65, 117),
(79, 110, 157),
(95, 110, 131),
(52, 54, 77),
(93, 98, 118),
(80, 87, 131),

(34, 35, 92),

(135, 151, 187),
(124, 144, 145),
(70, 91, 156),
(135, 153, 177),
(157, 166, 165),
(71, 82, 128),
(53, 69, 146),
(36, 53, 133),
(82, 107, 200),
(79, 123, 134),
(68, 100, 161),
(72, 110, 185),
(59, 97, 178),
(58, 86, 169),
(83, 103, 172),
(59, 90, 154),
(19, 39, 125),
(113, 143, 193),
(65, 101, 153),
(23, 42, 108),
(34, 47, 99),
(133, 149, 165),
(82, 93, 172),
(48, 66, 164),
(46, 69, 145),
(37, 52, 119),
(91, 106, 147),
(40, 57, 85),
(57, 68, 114),
(12, 23, 68),
(16, 30, 65),
(60, 76, 138),
(34, 50, 127),
(24, 39, 120),
(35, 48, 127),
(23, 33, 118),
(9, 15, 77),
(14, 21, 73),
(22, 30, 77),
(16, 24, 73),
(32, 40, 86),
(38, 50, 76),
(68, 81, 100),
(32, 42, 67),
(108, 117, 148),
(36, 43, 69),
(0, 2, 14),
(60, 61, 56),
(13, 17, 16),
(2, 5, 20),
(7, 10, 27),
(31, 31, 33),
(93, 98, 102),
(72, 75, 126),
(47, 55, 140),

(43, 57, 145),

(35, 53, 129),
(38, 56, 140),
(40, 58, 134),
(39, 53, 118),
(15, 28, 81),
(36, 49, 93),
(39, 51, 101),
(62, 72, 131),
(79, 92, 136),
(65, 82, 128),
(49, 66, 136),
(87, 108, 165),
(39, 57, 107),
(46, 61, 118),
(39, 54, 97),
(55, 68, 112),
(27, 40, 72),
(28, 36, 82),
(26, 36, 72),
(33, 46, 80),
(38, 48, 109),
(51, 66, 135),
(46, 67, 114),
(28, 49, 94),
(50, 71, 116),
(42, 66, 104),
(29, 51, 108),
(38, 58, 129),
(17, 32, 89),
(60, 73, 128),
(38, 49, 115),
(36, 51, 106),
(52, 70, 118),
(42, 65, 106),
(61, 83, 130),
(66, 104, 177),
(75, 105, 177),
(65, 95, 149),
(62, 99, 144),
(129, 149, 173),
(64, 76, 98),
(60, 73, 125),
(39, 55, 117),
(74, 91, 137),
(75, 100, 164),
(68, 106, 169),
(66, 107, 153),
(174, 200, 223),
(97, 108, 164),
(40, 48, 97),
(19, 31, 79),
(43, 60, 103),
(63, 82, 124),
(23, 45, 84),
(29, 41, 101),

(44, 54, 123),

(35, 48, 101),
(46, 61, 100),
(79, 94, 137),
(36, 57, 122),
(86, 114, 162),
(68, 87, 130),
(145, 156, 174),
(43, 58, 113),
(66, 95, 163),
(62, 94, 145),
(80, 106, 131),
(118, 139, 160),
(121, 140, 170),
(88, 109, 156),
(115, 136, 193),
(87, 121, 158),
(104, 127, 141),
(189, 192, 185),
(96, 100, 109),
(85, 115, 167),
(103, 138, 180),
(55, 77, 124),
(98, 107, 138),
(167, 172, 152),
(121, 131, 158),
(179, 191, 187),
(42, 45, 34),
(7, 8, 26),
(0, 3, 25),
(46, 71, 93),
(96, 128, 167),
(108, 131, 181),
(36, 54, 102),
(56, 70, 133),
(38, 53, 118),
(29, 47, 95),
(10, 31, 74),
(106, 131, 161),
(86, 110, 156),
(86, 116, 142),
(86, 114, 153),
(109, 141, 182),
(110, 144, 189),
(92, 123, 167),
(73, 96, 130),
(83, 99, 135),
(58, 73, 130),
(44, 61, 115),
(55, 74, 132),
(55, 76, 141),
(35, 55, 126),
(37, 67, 121),
(27, 53, 110),
(72, 97, 154),
(90, 111, 174),

(61, 81, 144),

(75, 90, 149),
(52, 76, 122),
(49, 60, 114),
(193, 199, 185),
(118, 129, 157),
(102, 124, 163),
(51, 75, 113),
(106, 124, 172),
(27, 44, 96),
(37, 51, 100),
(39, 56, 102),
(45, 65, 115),
(48, 70, 109),
(55, 75, 108),
(54, 74, 111),
(52, 69, 115),
(177, 172, 168),
(193, 180, 145),
(167, 148, 115),
(191, 176, 137),
(175, 159, 126),
(21, 22, 52),
(15, 17, 56),
(22, 26, 53),
(12, 15, 50),
(17, 19, 42),
(65, 67, 88),
(18, 20, 45),
(10, 15, 35),
(17, 25, 36),
(0, 9, 24),
(2, 13, 35),
(15, 23, 42),
(21, 31, 56),
(0, 10, 23),
(0, 3, 28),
(86, 103, 133),
(74, 105, 149),
(82, 110, 150),
(63, 85, 122),
(81, 96, 139),
(77, 89, 141),
(16, 35, 67),
(69, 88, 128),
(42, 64, 103),
(36, 54, 102),
(72, 85, 137),
(186, 198, 224),
(50, 65, 96),
(85, 102, 148),
(91, 115, 151),
(107, 133, 166),
(122, 141, 156),
(139, 147, 106),
(111, 138, 121),

(105, 146, 142),

(112, 141, 157),
(147, 160, 150),
(161, 167, 141),
(139, 151, 131),
(188, 201, 209),
(109, 124, 129),
(142, 158, 147),
(93, 109, 108),
(107, 126, 120),
(76, 95, 76),
(135, 154, 132),
(73, 94, 113),
(152, 179, 174),
(66, 81, 86),
(90, 106, 121),
(95, 116, 159),
(117, 140, 182),
(131, 153, 164),
(159, 189, 181),
(132, 152, 159),
(111, 124, 143),
(16, 22, 46),
(24, 28, 53),
(18, 23, 52),
(15, 19, 48),
(34, 38, 76),
(18, 24, 72),
(168, 180, 222),
(60, 78, 124),
(38, 56, 106),
(22, 33, 87),
(23, 30, 84),
(18, 22, 70),
(32, 34, 82),
(38, 40, 89),
(27, 29, 68),
(17, 19, 57),
(31, 35, 73),
(49, 52, 95),
(33, 45, 93),
(31, 40, 97),
(17, 24, 92),
(27, 39, 99),
(32, 49, 103),
(76, 111, 149),
(75, 112, 141),
(53, 87, 122),
(69, 103, 149),
(90, 125, 163),
(89, 119, 145),
(30, 46, 79),
(95, 127, 148),
(71, 101, 125),
(75, 95, 119),
(15, 31, 44),

(0, 17, 21),

(58, 87, 95),
(77, 108, 126),
(0, 12, 42),
(51, 45, 79),
(34, 45, 73),
(44, 54, 89),
(30, 42, 82),
(35, 49, 94),
(41, 55, 102),
(89, 102, 136),
(107, 117, 152),
(9, 16, 60),
(3, 9, 61),
(31, 38, 90),
(40, 50, 99),
(20, 33, 65),
(27, 32, 98),
(35, 43, 108),
(16, 26, 95),
(0, 14, 66),
(43, 61, 125),
(59, 76, 145),
(36, 49, 93),
(50, 58, 94),
(12, 26, 65),
(5, 31, 105),
(54, 69, 162),
(40, 54, 115),
(83, 111, 161),
(108, 140, 187),
(28, 48, 119),
(66, 88, 146),
(62, 83, 136),
(59, 80, 81),
(151, 177, 106),
(95, 116, 85),
(119, 135, 148),
(149, 167, 169),
(135, 153, 127),
(75, 90, 49),
(114, 134, 135),
(114, 127, 120),
(119, 136, 118),
(156, 173, 180),
(106, 130, 156),
(61, 91, 141),
(59, 95, 169),
(53, 87, 174),
(72, 101, 179),
(87, 120, 155),
(86, 112, 173),
(26, 49, 119),
(62, 80, 94),
(118, 140, 137),
(41, 66, 146),

(64, 77, 169),

(75, 98, 174),
(62, 77, 142),
(54, 71, 127),
(36, 53, 105),
(25, 46, 91),
(70, 83, 161),
(40, 59, 117),
(5, 30, 70),
(75, 103, 150),
(76, 104, 169),
(40, 64, 134),
(59, 88, 148),
(33, 67, 139),
(30, 53, 131),
(21, 40, 142),
(18, 42, 132),
(125, 146, 193),
(49, 67, 129),
(18, 33, 98),
(28, 44, 96),
(0, 5, 48),
(69, 86, 102),
(75, 94, 111),
(105, 128, 142),
(80, 110, 112),
(81, 110, 144),
(20, 51, 82),
(86, 104, 140),
(36, 54, 100),
(52, 69, 112),
(46, 62, 95),
(65, 87, 128),
(47, 70, 114),
(29, 33, 97),
(99, 119, 144),
(54, 76, 100),
(86, 99, 116),
(173, 181, 168),
(189, 202, 176),
(147, 162, 185),
(38, 49, 157),
(74, 93, 151),
(63, 79, 156),
(32, 49, 105),
(25, 32, 87),
(34, 42, 89),
(72, 94, 154),
(102, 127, 183),
(2, 18, 80),
(40, 55, 136),
(80, 97, 185),
(71, 98, 151),
(53, 84, 141),
(44, 71, 152),
(64, 95, 178),

(71, 97, 171),

(56, 76, 163),
(28, 47, 115),
(29, 43, 90),
(95, 109, 146),
(24, 29, 110),
(47, 55, 136),
(33, 48, 139),
(28, 41, 120),
(82, 92, 102),
(19, 24, 43),
(18, 17, 61),
(27, 27, 63),
(5, 10, 16),
(34, 43, 50),
(46, 64, 126),
(12, 29, 99),
(38, 49, 138),
(33, 41, 122),
(17, 28, 82),
(11, 15, 44),
(7, 14, 33),
(0, 2, 28),
(28, 37, 70),
(8, 19, 51),
(38, 51, 103),
(38, 51, 106),
(41, 51, 112),
(34, 41, 109),
(35, 40, 104),
(28, 38, 89),
(32, 46, 93),
(24, 41, 95),
(30, 44, 115),
(27, 43, 121),
(37, 49, 131),
(40, 57, 129),
(36, 51, 132),
(38, 54, 129),
(47, 63, 125),
(30, 42, 102),
(39, 48, 105),
(51, 54, 123),
(48, 50, 101),
(24, 30, 56),
(44, 50, 84),
(47, 62, 103),
(33, 54, 97),
...]
In [88]: image = tf.keras.preprocessing.image.img_to_array(image)
In [89]: print(type(image)) #now its in numpy

print(image.shape)

<class 'numpy.ndarray'>
(225, 375, 3)
In [90]: image.min(), image.max() # normalization is required
Out[90]: (0.0, 255.0)
In [91]: # image = image / 255

image = tf.keras.applications.inception_v3.preprocess_input(image)
In [92]: image.min(), image.max()
Out[92]: (-1.0, 1.0)
In [93]: image_batch = tf.expand_dims(image, axis = 0) #or np.expand_dims(image, axis = 0)
In [106… activations = deep_dream_model.predict(image_batch)
1/1 [==============================] - 0s 23ms/step
In [95]: len(activations)
Out[95]: 2
In [96]: deep_dream_model.outputs
Out[96]: [<KerasTensor: shape=(None, None, None, 768) dtype=float32 (created by layer 'mixe
d3')>,
<KerasTensor: shape=(None, None, None, 768) dtype=float32 (created by layer 'mixe
d5')>]
In [97]: # calculate loss

def calculate_loss(image, network):
image_batch = tf.expand_dims(image, axis = 0 )
activations = network.predict(image_batch)
losses = []
for act in activations:

loss = tf.math.reduce_mean(act)
losses.append(loss)
print(losses)
print(np.shape(losses))
print(tf.reduce_sum(losses))
In [98]: calculate_loss(image,deep_dream_model)
1/1 [==============================] - 0s 22ms/step

[<tf.Tensor: shape=(), dtype=float32, numpy=0.45276394>, <tf.Tensor: shape=(), dtype
=float32, numpy=0.16531265>]
(2,)
tf.Tensor(0.61807656, shape=(), dtype=float32)

In [99]: #Gradient Ascent

@tf.function # as if it were a global value
def deep_dream(network,image, learning_rate):

with tf.GradientTape() as tape: # GradientTape() its a tf class to access
the gradients
tape.watch(image)
loss = calculate_loss(image,network)
gradients = tape.gradient(loss,image ) #derivate
gradients /= tf.math.reduce_std(gradients)
In [100… # Compare the activations with the pixels

# Emphasize parts of the image
# Change the pixels of the input image
@tf.function
def deep_dream(network, image, learning_rate):
with tf.GradientTape() as tape:
tape.watch(image)
loss = calculate_loss(image, network)
gradients = tape.gradient(loss, image) # Derivate

gradients /= tf.math.reduce_std(gradients)
image = image + gradients * learning_rate
image = tf.clip_by_value(image, -1, 1)
return loss, image
In [101… def inverse_transform(image):

image = 255 * (image + 1.0) / 2.0
return tf.cast(image, tf.uint8)
In [102… def run_deep_dream(network, image, epochs, learning_rate):

for epoch in range(epochs):
loss, image = deep_dream(network, image, learning_rate)
if epoch % 200 == 0:
plt.figure(figsize=(12,12))
plt.imshow(inverse_transform(image))
plt.show()
print('Epoch {}, loss {}'.format(epoch, loss))
In [103… image.shape, type(image)
Out[103… ((225, 375, 3), numpy.ndarray)
In [107… run_deep_dream(network=deep_dream_model, image=image, epochs = 8000,

learning_rate=0.0001)

---------------------------------------------------------------------------
RuntimeError Traceback (most recent call last)
Cell In[107], line 1
----> 1 run_deep_dream(network=deep_dream_model, image=image, epochs = 8000, learnin
g_rate=0.0001)
Cell In[102], line 3, in run_deep_dream(network, image, epochs, learning_rate)

1 def run_deep_dream(network, image, epochs, learning_rate):
2 for epoch in range(epochs):
----> 3 loss, image = deep_dream(network, image, learning_rate)
5 if epoch % 200 == 0:
6 plt.figure(figsize=(12,12))
File ~\anaconda3\envs\DEEP_LEARNIGNG\lib\site-packages\tensorflow\python\util\traceb
ack_utils.py:153, in filter_traceback.<locals>.error_handler(*args, **kwargs)
151 except Exception as e:
152 filtered_tb = _process_traceback_frames(e.__traceback__)
--> 153 raise e.with_traceback(filtered_tb) from None
154 finally:
155 del filtered_tb
File ~\AppData\Local\Temp\__autograph_generated_fileebdocjr6.py:12, in outer_factor

y.<locals>.inner_factory.<locals>.tf__deep_dream(network, image, learning_rate)
10 with ag__.ld(tf).GradientTape() as tape:
11 ag__.converted_call(ag__.ld(tape).watch, (ag__.ld(image),), None, fscop
e)
---> 12 loss = ag__.converted_call(ag__.ld(calculate_loss), (ag__.ld(image), ag_
_.ld(network)), None, fscope)
13 gradients = ag__.converted_call(ag__.ld(tape).gradient, (ag__.ld(loss), ag_
_.ld(image)), None, fscope)
14 gradients = ag__.ld(gradients)
File ~\AppData\Local\Temp\__autograph_generated_filewc_871iz.py:9, in outer_factory.

<locals>.inner_factory.<locals>.tf__calculate_loss(image, network)
7 with ag__.FunctionScope('calculate_loss', 'fscope', ag__.STD) as fscope:
8 image_batch = ag__.converted_call(ag__.ld(tf).expand_dims, (ag__.ld(imag
e),), dict(axis=0), fscope)
----> 9 activations = ag__.converted_call(ag__.ld(network).predict, (ag__.ld(ima
ge_batch),), None, fscope)
10 losses = []
12 def get_state():
File ~\anaconda3\envs\DEEP_LEARNIGNG\lib\site-packages\keras\utils\traceback_utils.p
y:70, in filter_traceback.<locals>.error_handler(*args, **kwargs)
67 filtered_tb = _process_traceback_frames(e.__traceback__)
68 # To get the full stack trace, call:
69 # `tf.debugging.disable_traceback_filtering()`
---> 70 raise e.with_traceback(filtered_tb) from None
71 finally:
72 del filtered_tb
File ~\anaconda3\envs\DEEP_LEARNIGNG\lib\site-packages\keras\engine\training.py:395
8, in _disallow_inside_tf_function(method_name)
3949 if tf.inside_function():
3950 error_msg = (
3951 "Detected a call to `Model.{method_name}` inside a `tf.function`. "

3952 "`Model.{method_name} is a high-level endpoint that manages its "

(...)
3956 "`model(x)`."
3957 ).format(method_name=method_name)
-> 3958 raise RuntimeError(error_msg)
RuntimeError: in user code:
File "C:\Users\Admin\AppData\Local\Temp\ipykernel_25408\2211669788.py", line 9,

in deep_dream *
loss = calculate_loss(image, network)
File "C:\Users\Admin\AppData\Local\Temp\ipykernel_25408\1401329825.py", line 4,
in calculate_loss *
activations = network.predict(image_batch)
File "C:\Users\Admin\anaconda3\envs\DEEP_LEARNIGNG\lib\site-packages\keras\utils
\traceback_utils.py", line 70, in error_handler **
raise e.with_traceback(filtered_tb) from None
File "C:\Users\Admin\anaconda3\envs\DEEP_LEARNIGNG\lib\site-packages\keras\engin
e\training.py", line 3958, in _disallow_inside_tf_function
raise RuntimeError(error_msg)
RuntimeError: Detected a call to `Model.predict` inside a `tf.function`. `Model.

predict is a high-level endpoint that manages its own `tf.function`. Please move the
call to `Model.predict` outside of all enclosing `tf.function`s. Note that you can c
all a `Model` directly on `Tensor`s inside a `tf.function` like: `model(x)`.
In [ ]:

Code DD

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Code DD

Uploaded by

Copyright:

Available Formats

3/5/24, 1:01 AM Untitled

In [75]: import tensorflow as tf

In [76]: base_model = tf.keras.applications.InceptionV3(include_top = False, weights =

localhost:8889/nbconvert/html/Dropbox/Code Practice/CV masterclass/Untitled.ipynb?download=false 1/47

conv2d_188 (Conv2D) (None, None, None, 864 ['input_3[0][0]']

batch_normalization_188 (Batch (None, None, None, 96 ['conv2d_188[0]

activation_188 (Activation) (None, None, None, 0 ['batch_normalizati

conv2d_189 (Conv2D) (None, None, None, 9216 ['activation_188[0]

batch_normalization_189 (Batch (None, None, None, 96 ['conv2d_189[0]

activation_189 (Activation) (None, None, None, 0 ['batch_normalizati

conv2d_190 (Conv2D) (None, None, None, 18432 ['activation_189[0]

batch_normalization_190 (Batch (None, None, None, 192 ['conv2d_190[0]

activation_190 (Activation) (None, None, None, 0 ['batch_normalizati

max_pooling2d_8 (MaxPooling2D) (None, None, None, 0 ['activation_190[0]

conv2d_191 (Conv2D) (None, None, None, 5120 ['max_pooling2d_8

batch_normalization_191 (Batch (None, None, None, 240 ['conv2d_191[0]

localhost:8889/nbconvert/html/Dropbox/Code Practice/CV masterclass/Untitled.ipynb?download=false 2/47

activation_191 (Activation) (None, None, None, 0 ['batch_normalizati

conv2d_192 (Conv2D) (None, None, None, 138240 ['activation_191[0]

batch_normalization_192 (Batch (None, None, None, 576 ['conv2d_192[0]

activation_192 (Activation) (None, None, None, 0 ['batch_normalizati

max_pooling2d_9 (MaxPooling2D) (None, None, None, 0 ['activation_192[0]

conv2d_196 (Conv2D) (None, None, None, 12288 ['max_pooling2d_9

batch_normalization_196 (Batch (None, None, None, 192 ['conv2d_196[0]

activation_196 (Activation) (None, None, None, 0 ['batch_normalizati

conv2d_194 (Conv2D) (None, None, None, 9216 ['max_pooling2d_9

conv2d_197 (Conv2D) (None, None, None, 55296 ['activation_196[0]

batch_normalization_194 (Batch (None, None, None, 144 ['conv2d_194[0]

batch_normalization_197 (Batch (None, None, None, 288 ['conv2d_197[0]

activation_194 (Activation) (None, None, None, 0 ['batch_normalizati

activation_197 (Activation) (None, None, None, 0 ['batch_normalizati

localhost:8889/nbconvert/html/Dropbox/Code Practice/CV masterclass/Untitled.ipynb?download=false 3/47

average_pooling2d_18 (AverageP (None, None, None, 0 ['max_pooling2d_9

conv2d_193 (Conv2D) (None, None, None, 12288 ['max_pooling2d_9

conv2d_195 (Conv2D) (None, None, None, 76800 ['activation_194[0]

conv2d_198 (Conv2D) (None, None, None, 82944 ['activation_197[0]

conv2d_199 (Conv2D) (None, None, None, 6144 ['average_pooling2d

batch_normalization_193 (Batch (None, None, None, 192 ['conv2d_193[0]

batch_normalization_195 (Batch (None, None, None, 192 ['conv2d_195[0]

batch_normalization_198 (Batch (None, None, None, 288 ['conv2d_198[0]

batch_normalization_199 (Batch (None, None, None, 96 ['conv2d_199[0]

activation_193 (Activation) (None, None, None, 0 ['batch_normalizati

activation_195 (Activation) (None, None, None, 0 ['batch_normalizati

activation_198 (Activation) (None, None, None, 0 ['batch_normalizati

activation_199 (Activation) (None, None, None, 0 ['batch_normalizati

mixed0 (Concatenate) (None, None, None, 0 ['activation_193[0]

localhost:8889/nbconvert/html/Dropbox/Code Practice/CV masterclass/Untitled.ipynb?download=false 4/47

conv2d_203 (Conv2D) (None, None, None, 16384 ['mixed0[0][0]']

batch_normalization_203 (Batch (None, None, None, 192 ['conv2d_203[0]

activation_203 (Activation) (None, None, None, 0 ['batch_normalizati

conv2d_201 (Conv2D) (None, None, None, 12288 ['mixed0[0][0]']

conv2d_204 (Conv2D) (None, None, None, 55296 ['activation_203[0]

batch_normalization_201 (Batch (None, None, None, 144 ['conv2d_201[0]

batch_normalization_204 (Batch (None, None, None, 288 ['conv2d_204[0]

activation_201 (Activation) (None, None, None, 0 ['batch_normalizati

activation_204 (Activation) (None, None, None, 0 ['batch_normalizati

average_pooling2d_19 (AverageP (None, None, None, 0 ['mixed0[0][0]']

conv2d_200 (Conv2D) (None, None, None, 16384 ['mixed0[0][0]']

conv2d_202 (Conv2D) (None, None, None, 76800 ['activation_201[0]

conv2d_205 (Conv2D) (None, None, None, 82944 ['activation_204[0]

conv2d_206 (Conv2D) (None, None, None, 16384 ['average_pooling2d

localhost:8889/nbconvert/html/Dropbox/Code Practice/CV masterclass/Untitled.ipynb?download=false 5/47

batch_normalization_200 (Batch (None, None, None, 192 ['conv2d_200[0]