jovsa

from pdb import set_trace
from tqdm import tqdm
from heapq import *

class Node:
    def __init__(self, val, next=None):
        self.val = val
        self.next = next

jovsa

from functools import lru_cache
from typing import Dict, List, NamedTuple, Optional, Tuple
from pdb import set_trace
import numpy as np


class Config(NamedTuple):
    idx: int
    rotation: int  # 0, 1, 2, 3
    flipped: bool

jovsa

# Topic: My Notes for Neural Machine Translation with Attention
# source: https://colab.research.google.com/github/tensorflow/tensorflow/blob/master/tensorflow/contrib/eager/python/examples/nmt_with_attention/nmt_with_attention.ipynb#scrollTo=yJ_B3mhW3jFk

def gru(units):
  # If you have a GPU, we recommend using CuDNNGRU(provides a 3x speedup than GRU)
  # the code automatically does that.
  # You could also use an RNN or LSTM based on your requirements
  if tf.test.is_gpu_available():
    return tf.keras.layers.CuDNNGRU(units,
                                    return_sequences=True,

jovsa

Here are the steps to work initialize a virtual environment:
pip install virtualenv
$ pip install virtualenv                     // This installs virtual environment  //for python3 >  pip3 install virtualenv
$ cd my_project_folder      
$ python -m virtualenv --no-site-packages venv            //Initialization  //for python3 >  python3 -m virtualenv .venv
$ source venv/bin/activate  // activates it
$ pip install -r requirements.txt
$ pip freeze > requirements.txt (if there is no requirements file)
     OR
$ pip freeze >> requirements.txt (if a requirements file exists and you want to override it)

jovsa

IMPORTANT NOTE: The numbers reported below are just estimates that are meant to be used for back-of-the-envelope calculations
========================================================================================================================================
Software Engineering Numbers (~2019)
========================================================================================================================================

Cache/Lookup
----------------------------------
L1 cache reference                         0.5   ns
L2 cache reference                           4   ns                        8x L1 cache
Main memory reference                      100   ns                        25x L2 cache, 200x L1 cache

jovsa

"""Ensembling Script: Implementing a regression stacker.
A ensembling script that can stack models from XGBoost and Sklearn. 
Inspiration has been drawn from Emanuele Olivetti (emanuele) 
[link: https://github.com/emanuele/kaggle_pbr/blob/master/blend.py]. 

This script uses a K-fold split to train and predicts out-of-fold the blended datasets. 
The first half contains function definition  and the second half contains a functions call. 
Use this script as a framework and tweak it to fit your needs.
"""

jovsa

# Bagging Algorithm on the Sonar dataset (ref: http://machinelearningmastery.com/implement-bagging-scratch-python/)
from random import seed
from random import randrange
from csv import reader

# Load a CSV file
def load_csv(filename):
	dataset = list()
	with open(filename, 'r') as file:
		csv_reader = reader(file)

jovsa

from csv import reader
import numpy as np
from random import seed
from sklearn.ensemble import BaggingClassifier, RandomForestClassifier
from sklearn.model_selection import cross_val_score

## Pre-processing functions
# Load a CSV file
def load_csv(filename):
	dataset = list()