parajain

Latency Comparison Numbers (~2012)
----------------------------------
L1 cache reference                           0.5 ns
Branch mispredict                            5   ns
L2 cache reference                           7   ns                      14x L1 cache
Mutex lock/unlock                           25   ns
Main memory reference                      100   ns                      20x L2 cache, 200x L1 cache
Compress 1K bytes with Zippy             3,000   ns        3 us
Send 1K bytes over 1 Gbps network       10,000   ns       10 us
Read 4K randomly from SSD*             150,000   ns      150 us          ~1GB/sec SSD

parajain

import nltk
#nltk.download('omw-1.4')
import tqdm
from nltk.corpus import wordnet as wn

all_nouns = [word for synset in wn.all_synsets('n') for word in synset.lemma_names()]

inputphrase= ''
wordlens = [len(w) for w in inputphrase.split()]
t=0

parajain

class GlobalMaxPooling1D(nn.Module):
    '''
    https://keras.io/api/layers/pooling_layers/global_max_pooling1d/
    Code: https://discuss.pytorch.org/t/equivalent-of-keras-globalmaxpooling1d/45770/5
    Input:
    * If data_format='channels_last': 3D tensor with shape: (batch_size, steps, features)
    * If data_format='channels_first': 3D tensor with shape: (batch_size, features, steps)
    Output:
    * 2D tensor with shape (batch_size, features).
    '''

parajain

from flair.data import Sentence
from flair.models import SequenceTagger
import sys

class FlairChunker():
    def __init__(self):
        self.chunker = SequenceTagger.load('chunk')

    def get_chunk_spans(self, s):
        sentence = Sentence(s)

parajain

'''
https://www.clips.uantwerpen.be/conll2001/clauses/
Clauses are word sequences which contain a subject and a predicate. Here is an example of a sentence and its clauses obtained from Wall Street Journal section 15 of the Penn Treebank [MSM93]:

   (S The deregulation of railroads and trucking companies
      (SBAR that
          (S began in 1980)
      ) 
      enabled
      (S shippers to bargain for transportation)

parajain

#Download bert from command line

wget --load-cookies /tmp/cookies.txt "https://docs.google.com/uc?export=download&confirm=$(wget --quiet --save-cookies /tmp/cookies.txt --keep-session-cookies --no-check-certificate 'https://docs.google.com/uc?export=download&id=1f_LEWVgrtZLRuoiExJa5fNzTS8-WcAX9' -O- | sed -rn 's/.*confirm=([0-9A-Za-z_]+).*/\1\n/p')&id=1f_LEWVgrtZLRuoiExJa5fNzTS8-WcAX9" -O pytorch_model_uncased_L-12_H-768_A-12.bin && rm -rf /tmp/cookies.txt

parajain

"""
Example Generate basic documeentation for arguments. :)
python generate_doc.py -md > doc.md
"""
import argparse

def add_md_help_argument(parser):
    """ md help parser """
    parser.add_argument('-md', action=MarkdownHelpAction,
                        help='print Markdown-formatted help text and exit.')

parajain

import numpy as np

def log_softmax(x):
    e_x = np.exp(x - np.max(x))
    return np.log(e_x / e_x.sum())

def lognormalize(x):
    a = np.logaddexp.reduce(x)
    return np.exp(x - a)
    

parajain

'''
Example to print flag as key,values. So that can be saved as json.
tf version '1.10.0'
'''

import tensorflow as tf

tf.app.flags.DEFINE_string('source_vocabulary', 'data/europarl-v7.1.4M.de.json', 'Path to source vocabulary')
tf.app.flags.DEFINE_string('target_vocabulary', 'data/europarl-v7.1.4M.fr.json', 'Path to target vocabulary')

parajain

'''
Basic text data cleaning script
Tokenization, remove punctuation
'''

import sys
import re
import string
from nltk.tokenize import word_tokenize