joaopcnogueira’s gists

joaopcnogueira / how_to_build_regressive_features.R

Last active November 3, 2023 02:35

Piece of code demonstrating how to build regressive features for machine learning modeling, such as the sum of sales for the last 3 months, the mean of sales for the last 6 months and so on.

	# Creating a spine table with three columns ----
	# customer_id: identifier of the customer, for which we are going to predict the next month sales
	# year_month: reference date
	# sales: the metric we want to predict
	spine_tbl <- tibble(
	customer_id = c(rep("João", 24), rep("Denise", 24)),
	year_month = c( seq( ymd("2021-11-01"), ymd("2023-10-01"), by = '1 month' ), seq( ymd("2021-11-01"), ymd("2023-10-01"), by = '1 month' ) ),
	sales = sample(100:1000, 48, replace = TRUE)
	)

joaopcnogueira / lpsolve.R

Last active November 15, 2021 15:44

Gist Example of Optimization and Linear Programming with R

	# Import lpSolve package
	library(lpSolve)

	#
	# Set up the problem: maximize
	# z = 2x1 + 11x2 subject to
	# 2x1 + 2x2 <= 20
	# x1 + 2*x2 <= 12
	# 3x1 + 4x2 <= 36
	# x1 <= 5

joaopcnogueira / python_project.py

Created November 29, 2019 03:54

How to make Python Code Run Independent of OS Path

	# Work Directory (/home/user/python_project/):
	# - data
	# - data/employee.csv
	# - src
	#

	import os

	# WORK_DIR="/home/user/python_project/"
	WORK_DIR = os.getcwd()

joaopcnogueira / condaenv.txt

Created August 20, 2019 13:21 — forked from pratos/condaenv.txt

To package a conda environment (Requirement.txt and virtual environment)

	# For Windows users# Note: <> denotes changes to be made

	#Create a conda environment
	conda create --name <environment-name> python=<version:2.7/3.5>

	#To create a requirements.txt file:
	conda list #Gives you list of packages used for the environment

	conda list -e > requirements.txt #Save all the info about packages to your folder

joaopcnogueira / target_mean_encoder.R

Last active September 22, 2019 23:53

Target mean encoder implementation in R

	library(dplyr)

	# creating a toy dataset
	data = tibble(vehicle = c("car", "bus", "bike", "bus", "car", "bike"),
	target = c(23,34,56,78,33,65))

	# print dataframe
	data

	# OUTPUT

joaopcnogueira / titanic-pipeline4.py

Last active July 11, 2019 21:02

Using Pipelines and ColumnTransform to compose different data pre-processing steps

	import pandas as pd
	from sklearn.tree import DecisionTreeClassifier
	from sklearn.model_selection import train_test_split
	from sklearn.pipeline import Pipeline
	from sklearn.impute import SimpleImputer
	from category_encoders import OneHotEncoder
	from sklearn.model_selection import KFold
	from sklearn.model_selection import cross_validate
	from sklearn.model_selection import GridSearchCV
	from sklearn.compose import ColumnTransformer

joaopcnogueira / titanic-pipeline3.py

Last active July 11, 2019 17:41

GridSearchCV with pipelines

	import pandas as pd
	from sklearn.tree import DecisionTreeClassifier
	from sklearn.model_selection import train_test_split
	from sklearn.pipeline import Pipeline
	from sklearn.impute import SimpleImputer
	from category_encoders import OneHotEncoder
	from sklearn.model_selection import KFold
	from sklearn.model_selection import cross_validate
	from sklearn.model_selection import GridSearchCV

joaopcnogueira / titanic-pipeline2.py

Last active July 11, 2019 14:55

K-fold cross-validation with pipeline

	import pandas as pd
	from sklearn.tree import DecisionTreeClassifier
	from sklearn.model_selection import train_test_split
	from sklearn.pipeline import Pipeline
	from sklearn.impute import SimpleImputer
	from category_encoders import OneHotEncoder
	from sklearn.model_selection import KFold
	from sklearn.model_selection import cross_validate

	# lendo o dataset

joaopcnogueira / titanic-pipeline.py

Last active July 11, 2019 14:08

Refactored titanic code with pipelines

	import pandas as pd
	from sklearn.tree import DecisionTreeClassifier
	from sklearn.model_selection import train_test_split
	from sklearn.pipeline import Pipeline
	from sklearn.impute import SimpleImputer
	from category_encoders import OneHotEncoder

	# lendo o dataset
	df = pd.read_csv("train.csv")

joaopcnogueira / groupkfold_example.py

Last active July 11, 2019 13:53

Simple Example using GroupKFold with Cross-Validate

	import numpy as np
	import pandas as pd
	from sklearn import datasets
	from sklearn.tree import DecisionTreeClassifier
	from sklearn.metrics import accuracy_score
	from sklearn.model_selection import GroupKFold

	# Loading the data
	iris = datasets.load_iris()
	design_matrix = np.concatenate((iris['data'], iris['target'].reshape(150,1)), axis=1)

João Paulo Nogueira joaopcnogueira