Intro
R statistical analysis, statistical support
Python general data science.
R > functional, Python > object-oriented.
R > data analysis functionality built-in, Python relies on packages.
Python > non-statistical tasks.
Both can handle huge size of database.
Python is faster, better for deep learning.
R is better for data visualization.
Resources:
main, definition, comparison |
|
|
R
Importing a CSV, Data Look
library(readr)
nba <- read_csv("nba_2013.csv")
dim(nba)
head(nba, 1)
Averages for Each Statistic
library(purrr)
library(dplyr)
nba %>%
select_if(is.numeric) %>%
map_dbl(mean, na.rm = TRUE)
Scatterplots (see below results)
library(GGally)
nba %>%
select(ast, fg, trb) %>%
gpairs()
Data into Training and Testing Sets
trainRowCount <- floor(0.8 * nrow(nba))
set.seed(1)
trainIndex <- sample(1:nrow(nba),
+trainRowCount)
train <- nba[trainIndex,]
test <- nba[-trainIndex,]
Univariate Linear Regression
fit <- lm(ast ~ fg, data=train)
predictions <- predict(fit, test)
Summary Statistics
summary(fit)
Web Scrapping
library(RCurl)
url <- "http"
data <- readLines(url)
|
|
|
Python
Importing a CSV, Data Look
import pandas
nba = pandas.read_csv
("nba_2013.csv")
nba.shape
nba.head(1)
Averages for Each Statistic
nba.mean()
Scatterplots (see below results)
import seaborn as sns
span class="token keyword"
import matplotlib.pyplot as plt
sns.pairplot(nba[["ast", "fg", "trb"]])
plt.show()
Data into Training/Testing Set
train = nba.sample(frac=0.8, random_state=1)
test = nba.loc[~nba.index.isin
(train.index)]
Univariate Linear Regression
fit <- lm(ast ~ fg, data=train)
predictions <- predict(fit, test)
Summary Statistics
import statsmodels.formula.api
as sm
model = sm.ols(formula='ast ~ fga'
,data=train)
fitted = model.fit()
fitted.summary()
Web Scrapping
import requests
url = "http"
data = requests.get(url).content
|
|
