from dagbldr.datasets import fetch_binarized_mnist, minibatch_iterator
from dagbldr.utils import get_params
from dagbldr.utils import TrainingLoop
from dagbldr.utils import create_checkpoint_dict
from dagbldr.nodes import softplus
from dagbldr.nodes import sigmoid
from dagbldr.nodes import linear
from dagbldr.nodes import gaussian_log_sample
from dagbldr.nodes import gaussian_log_kl
from dagbldr.nodes import binary_crossentropy
from dagbldr.optimizers import adam
mnist = fetch_binarized_mnist()
X = mnist["data"].astype("float32")
X_sym = tensor.fmatrix()
# random state so script is deterministic
random_state = np.random.RandomState(1999)
minibatch_size = 100
n_code = 100
n_hid = 200
width = 28
height = 28
n_input = width * height
# encode path aka q
from collections import OrderedDict import numpy as np import theano from dagbldr.datasets import fetch_binarized_mnist, minibatch_iterator from dagbldr.optimizers import adam from dagbldr.utils import add_datasets_to_graph, get_params_and_grads from dagbldr.utils import convert_to_one_hot, create_or_continue_from_checkpoint_dict from dagbldr.utils import TrainingLoop from dagbldr.nodes import softplus_layer, linear_layer, sigmoid_layer from dagbldr.nodes import gaussian_log_sample_layer, gaussian_log_kl from dagbldr.nodes import binary_crossentropy, softmax_layer from dagbldr.nodes import categorical_crossentropy mnist = fetch_binarized_mnist() train_indices = mnist["train_indices"] train_end = len(train_indices) valid_indices = mnist["valid_indices"] X = mnist["data"] y = mnist["target"] n_targets = 10 y = convert_to_one_hot(y, n_targets) # graph holds information necessary to build layers from parents graph = OrderedDict() X_sym, y_sym = add_datasets_to_graph([X, y], ["X", "y"], graph) # random state so script is deterministic random_state = np.random.RandomState(1999) minibatch_size = 100 n_code = 100