duttashi

# Method 1: Easiest and does not require any library
data(mtcars)

## 75% of the sample size
smp_size <- floor(0.75 * nrow(mtcars))

## set the seed to make your partition reproducible
set.seed(123)
train_ind <- sample(seq_len(nrow(mtcars)), size = smp_size)

duttashi

# Apparently the problem lies in https. The function read.csv() in R fails at this. I tried RCurl's getURL() still same error.
# Then I tried fread() from library(data.table) and it worked.
# I give below a minimum reproducible example to download data from a https base webpage.

# load the adult dataset
library(data.table)
dt<- fread("https://archive.ics.uci.edu/ml/machine-learning-databases/adult/adult.data",header = FALSE, sep=",")

head(dt)
   V1               V2     V3        V4 V5                 V6

duttashi

# Ensure the data is read as a dataframe and that the categorical variables are read as factors and not characters.
# A minimum reprex is given below

# load the adult dataset from the UCI ML repo.
library(data.table)
dt<- fread("https://archive.ics.uci.edu/ml/machine-learning-databases/adult/adult.data",
           header = FALSE, sep = ",", stringsAsFactors = TRUE)
# coerce data table to data frame
dt<- as.data.frame(dt)
head(dt)

duttashi

# create function to plot linear regression results
# adapted from https://sejohnston.com/2012/08/09/a-quick-and-easy-function-to-plot-lm-results-in-r/
ggplotRegression <- function (fit) {
  lmdf<- data.frame(fitted_values = fit$fitted.values, actual_values = fit$model[, 1])
  print(names(lmdf))
  ggplot(lmdf, aes(x = actual_values, y = fitted_values)) +
    geom_point() +
    geom_abline(slope = 1, intercept = 0) +
    labs(title = paste("Adj R2 = ", signif(summary(fit)$adj.r.squared, 4),
                       "Intercept =",signif(fit$coef[[1]],5 ),

duttashi

There 3 options how to convert categorical features to numerical:

• Use OneHotEncoder. You will transform categorical feature to four new columns, where will be just one 1 and other 0. The problem here is that difference between "morning" and "afternoon" is the same as the same as "morning" and "evening".

• Use OrdinalEncoder. You transform categorical feature to just one column. "morning" to 1, "afternoon" to 2 etc. The difference between "morning" and "afternoon" will be smaller than "morning" and "evening" which is good, but the difference between "morning" and "night" will be greatest which might not be what you want.

• Use transformation that I call two_hot_encoder. It is similar to OneHotEncoder, there are just two 1 in the row. The difference between The difference between "mor

duttashi

# load required libraries
library(tidyverse)
# READ DATA IN MEMORY
df_train<- read.csv("kaggle_fake_job_prediction/data/fake_job_postings.csv",
                    header=T, na.strings=c(" ","NA"), stringsAsFactors = FALSE, strip.white = TRUE)

# create copy
df<- df_train
df<- df %>%
  mutate_if(is.character, funs(factor(.)))

duttashi

# load required libraries
library(tidyverse)
# READ DATA IN MEMORY
df_train<- read.csv("kaggle_fake_job_prediction/data/fake_job_postings.csv",
                    header=T, na.strings=c(" ","NA"), stringsAsFactors = FALSE, strip.white = TRUE)

# create copy
df<- df_train
# coerce character vars to factor for data cleanup
df<- df %>%

duttashi

# create a custom function to split data into 3 sets

import numpy as np

def train_validate_test_split(df, train_percent=.6, validate_percent=.2, seed=None):
    np.random.seed(seed)
    perm = np.random.permutation(df.index)
    m = len(df.index)
    train_end = int(train_percent * m)
    validate_end = int(validate_percent * m) + train_end

duttashi

Problem statement

Did some data analysis which resulted in generating a huge data file, greater than 100 MB. Accidentaly, tried to push it to Github and the nightmares began! Keep getting error messages, cant push because large files detected.

Solution

The following solution worked for me;

1. Open command shell in the repo

duttashi

# suppose the text data is loaded in a dataframe called, df.
# using regular expressions to clean the text data

#Remove twitter handlers
df.text = df.text.apply(lambda x:re.sub('@[^\s]+','',x))

#remove hashtags
df.text = df.text.apply(lambda x:re.sub(r'\B#\S+','',x))

# Remove URLS