BrianPugh

Create a tmux conf file
#> touch ~/.tmux.conf

Install TMP
#> git clone https://github.com/tmux-plugins/tpm ~/.tmux/plugins/tpm

[Copy and paste tmux.conf below into local file.]

Load tmux configurations
#> tmux source-file ~/.tmux.conf

BrianPugh

def get_checkpoints(checkpoint_dir):
    '''
    Finds all checkpoints in a directory and returns them in order
    from least iterations to most iterations
    '''
    meta_list=[]
    for file in os.listdir(checkpoint_dir):
        if file.endswith('.meta'):
            meta_list.append(os.path.join(checkpoint_dir, file[:-5]))
    meta_list = sort_nicely(meta_list)

BrianPugh

def optimistic_restore(session, save_file, variable_scope=''):
    '''
    A Caffe-style restore that loads in variables
    if they exist in both the checkpoint file and the current graph.
    Call this after running the global init op.
    By DanielGordon10 on December 27, 2016
    https://github.com/tensorflow/tensorflow/issues/312
    With RalphMao tweak.

    bpugh, July 21, 2017: Added a variable_scope so that a network can be

BrianPugh

#!/user/bin/env python

'''
This is a template file to get your project on the road quickly!

Last updated: July 27th, 2017
'''

__author__  = 'Brian Pugh'
__email__   = 'bnp117@gmail.com'

BrianPugh

BrianPugh

BrianPugh

# Recursively copy all the avi videos in some file structure to a flat folder
# Source: https://stackoverflow.com/a/1936214
find . -iname '*.avi' -exec cp \{\} ../videos/ \;

# ffmpeg convert avi to mp4
ffmpeg -i infile.avi youroutput.mp4

# Convert all avi's in a folder to mp4 with ffmpeg
# Source: https://stackoverflow.com/a/33766147
for i in *.avi; do ffmpeg -i "$i" -vf "scale=trunc(iw/2)*2:trunc(ih/2)*2" "${i%.*}.mp4"; done

BrianPugh

04fcd48e8946c42e7efa9fee5bde9ce58729b8bc389e9f280cd1f50ca8b262e29b8b59ef918f41617cef8d1bc40db64ce9d825aab81bbb39f02fe26a781a794bdd;bradennapier

BrianPugh

NEP-01: State Blocks

The implementation of state blocks (known as "universal blocks") is a protocol improvement that encodes all account data in every transaction. This allows the following:

1. Simplifies account balance computation and block verification
2. Enables inexpensive hardware to easily and securely sign transactions
3. Aggressive pruning of the block-lattice

Legacy Blocks

BrianPugh

# Name of all Variables in Graph
# Source: https://stackoverflow.com/questions/36883949/in-tensorflow-get-the-names-of-all-the-tensors-in-a-graph
from pprint import pprint
pprint([n.name for n in tf.get_default_graph().as_graph_def().node if "Variable" in n.op ])

# check for substring in list of strings
print( "\n".join(s for s in list if sub in s) )