ikhlestov

def mi_linear(arg1, arg2, output_size, global_bias_start=0.0, scope=None):
    """Multiplicated Integrated Linear map:
    See http://arxiv.org/pdf/1606.06630v1.pdf
    A * (W[0] * arg1) * (W[1] * arg2) + (W[0] * arg1 * bias1) + (W[1] * arg2 * bias2) + global_bias.
    Args:
        arg1: batch x n, Tensor.
        arg2: batch x n, Tensor.
        output_size: int, second dimension of W[i].
    global_bias_start: starting value to initialize the global bias; 0 by default.
        scope: VariableScope for the created subgraph; defaults to "MILinear".

ikhlestov

tmux shortcuts & cheatsheet

start new:

tmux

start new with session name:

tmux new -s myname

ikhlestov

import hashlib
import argparse
import subprocess

parser = argparse.ArgumentParser()
parser.add_argument('--md5', type=str, required=True)
args = parser.parse_args()

script_to_run = 'some.py'

ikhlestov

import json

import tensorflow as tf
from tensorflow.python.client import timeline

x = tf.random_normal([1000, 1000])
y = tf.random_normal([1000, 1000])
res = tf.matmul(x, y)

# Run the graph with full trace option

ikhlestov

def composite_function(self, _input, out_features, kernel_size=3):
    """Function from paper H_l that performs:
    - batch normalization
    - ReLU nonlinearity
    - convolution with required kernel
    - dropout, if required
    """
    with tf.variable_scope("composite_function"):
        # BN
        output = self.batch_norm(_input)

ikhlestov

import tensorflow as tf

class MultithreadedTensorProvider():

    """ A class designed to provide tensors input in a
    separate threads. """

    def __init__(self, capacity, sess, dtypes, shuffle_queue=False,
        number_of_threads=1):

ikhlestov

import tensorflow as tf
from tensorflow.python.client import timeline

a = tf.random_normal([2000, 5000])
b = tf.random_normal([5000, 1000])
res = tf.matmul(a, b)

with tf.Session() as sess:
    # add additional options to trace the session execution
    options = tf.RunOptions(trace_level=tf.RunOptions.FULL_TRACE)

ikhlestov

import os
import tempfile

import tensorflow as tf
from tensorflow.contrib.layers import fully_connected as fc
from tensorflow.examples.tutorials.mnist import input_data
from tensorflow.python.client import timeline

batch_size = 100

ikhlestov

import json


class TimeLiner:
    _timeline_dict = None

    def update_timeline(self, chrome_trace):
        # convert crome trace to python dict
        chrome_trace_dict = json.loads(chrome_trace)
        # for first run store full trace

ikhlestov

import torch
import numpy as np

numpy_tensor = np.random.randn(10, 20)

# convert numpy array to pytorch array
pytorch_tensor = torch.Tensor(numpy_tensor)
# or another way
pytorch_tensor = torch.from_numpy(numpy_tensor)