Skip to content

Instantly share code, notes, and snippets.

View joelouismarino's full-sized avatar

Joe Marino joelouismarino

View GitHub Profile
@joelouismarino
joelouismarino / googlenet.py
Last active October 24, 2024 05:51
GoogLeNet in Keras
from __future__ import print_function
import imageio
from PIL import Image
import numpy as np
import keras
from keras.layers import Input, Dense, Conv2D, MaxPooling2D, AveragePooling2D, ZeroPadding2D, Dropout, Flatten, Concatenate, Reshape, Activation
from keras.models import Model
from keras.regularizers import l2
from keras.optimizers import SGD
import numpy as np
def whiten(X, method='zca'):
"""
Whitens the input matrix X using specified whitening method.
Inputs:
X: Input data matrix with data examples along the first dimension
method: Whitening method. Must be one of 'zca', 'zca_cor', 'pca',
import math
import torch
from torch.distributions import Normal
# standard univariate Gaussian (Normal)
mean = torch.zeros(1)
std = torch.ones(1)
# evaluate from -0.5 to 0.5
x_min = -0.5 * torch.ones(1)
import math
import torch
from torch.distributions import Normal
# standard univariate Gaussian (Normal)
mean = torch.zeros(1)
std = torch.ones(1)
# evaluate at the origin
value = torch.zeros(1)
import numpy as np
n_samples = 500
mean_1 = 15
std_dev_1 = 5
mean_2 = -20
std_dev_2 = 3
X = np.concatenate([np.random.normal(mean_1, std_dev_1, n_samples / 2),
np.random.normal(mean_2, std_dev_2, n_samples / 2)], axis=0)
import numpy as np
n_samples = 500
mean = 15
std_dev = 5
X = np.random.normal(mean, std_dev, n_samples)
Z = (X - np.mean(X)) / np.std(X)
keras_top_inds = keras_act[0].argsort()[::-1][:5]
zip(keras_act[0][keras_top_inds], labels[keras_top_inds])
labels_file = caffe_root + 'data/ilsvrc12/synset_words.txt'
labels = np.loadtxt(labels_file, str, delimiter='\t')
caffe_top_inds = caffe_act[0].argsort()[::-1][:5]
zip(caffe_act[0][caffe_top_inds], labels[caffe_top_inds])
caffe_act = net.blobs[layer_name].data
layer = googlenet.get_layer(name=layer_name)
keras_act = get_activations(googlenet, layer, img)
import theano
def get_activations(model, layer, X_batch):
get_activations = theano.function([model.layers[0].input,K.learning_phase()], layer.output, allow_input_downcast=True)
activations = get_activations(X_batch,0)
return activations