---
title: "Quantified Self"
author: "Ashwin Mendhi"
output:
flexdashboard::flex_dashboard:
orientation: rows
vertical_layout: fill
source_code: embed
---
```{r setup, include=FALSE}
library(flexdashboard)
library(readxl)
library(ggplot2)
library(leaflet)
library(leaflet.extras)
library(ggmap)
```
About
=======================================================================
#### ANLY 512 Data Visualization Project
#### The Quantified Self
#### Ashwin Mendhi
I selected the folowing data sets to quantify and visualize my activity
**1. Daily Steps Count**
Data Collection - iPhone Health App. I used the Quantified Self app tp export the steps data from iPhone Health App. I then added travel and class schedule data from looking at my calender.
**2. Macros Data**
Data Collection - I used MyFitnessPal app on my phone to log my daily food intake. The app had an option to export this data and analysis purposes.
**3. September 2016 Location Data**
Data Collection - I downloaded my Google Location history and used this data to plot where I have been in September 2016 - my first month in the US. I wanted to track a more recent month/year but had stopped giving access to Google to track my loction from more than a year now. So selected September 2016 for analysis.
Steps
=======================================================================
Row {data-height=650}
-----------------------------------------------------------------------
### Travel
```{r}
stepsData <- read_xlsx("StepsData.xlsx", col_types = c("date", "date", "numeric", "numeric",
"numeric", "numeric", "numeric", "numeric",
"numeric", "numeric", "numeric", "numeric",
"guess", "text", "guess", "guess", "text"))
stepsData$Diet <- as.factor(stepsData$Diet)
steps <- stepsData[,c(1,10,12:16)]
steps <- subset(steps, Start >= "2020-07-31")
avgSteps <- mean(steps$`Steps (count)`)
stepsTrvl <- subset(steps, steps$`Travel outside Pittsburgh`== TRUE)
stepsNoTrvl <- subset(steps, steps$`Travel outside Pittsburgh`== FALSE)
avgStepsTrvl <- mean(stepsTrvl$`Steps (count)`)
avgStepsNoTrvl <- mean(stepsNoTrvl$`Steps (count)`)
ggplot(data = steps) +
geom_col(data = steps, aes(x=Start, y=`Steps (count)`, fill = `Travel outside Pittsburgh`)) +
labs(x="Day", y="Steps per day", fill=" " ) + lims(y=c(0,30000)) +
scale_fill_manual(values = c("grey30", "orange"),
breaks=c("FALSE", "TRUE"),
labels=c("No Travel Days", "Travel outside Pittsburgh")) +
theme_classic() + theme(legend.position = "bottom", text=element_text(size=18)) +
geom_hline(yintercept=avgStepsNoTrvl, color = "grey30", size = 1.5) +
geom_hline(yintercept=avgStepsTrvl, color = "orange", size=2) +
geom_label(aes(x=Start[1],y=avgStepsNoTrvl,label = round(avgStepsNoTrvl, digits = 0),
vjust=-0.5), hjust=-0.01, size=5, color="grey30") +
geom_label(aes(x=Start[1],y=avgStepsTrvl,label = round(avgStepsTrvl, digits = 0),
vjust=-0.5), hjust=-0.01, size=5, color="orange")
```
### HU Classes and Deliverable
```{r}
steps2 <- subset(steps, steps$Start >= "2020-07-31")
stepsClass <- subset(steps2, steps2$Huclass == TRUE | steps2$Hudeliverable == TRUE)
stepsNoClass <- subset(steps2, steps2$Huclass == FALSE & steps2$Hudeliverable == FALSE)
avgStepsClass <- mean(stepsClass$`Steps (count)`)
avgStepsNoClass <- mean(stepsNoClass$`Steps (count)`)
ggplot(data = steps2) +
geom_col(data = steps2, aes(x=Start, y=`Steps (count)`, fill = Huclass | Hudeliverable)) +
labs(x="Day", y="Steps per day", fill=" " ) + lims(y=c(0,30000)) +
scale_fill_manual(values = c("grey30", "#199787"),
breaks=c("FALSE", "TRUE"),
labels=c("No Class/Deliverable", "HU Class/Deliverable")) +
theme_classic() + theme(legend.position = "bottom", text=element_text(size=18)) +
geom_hline(yintercept=avgStepsNoClass, color = "grey30", size=1.5) +
geom_hline(yintercept=avgStepsClass, color = "#199787", size=2) +
geom_label(aes(x=Start[1],y=avgStepsNoClass,label = round(avgStepsNoClass, digits = 0),
vjust=-0.5), hjust=-0.01, size=5, color="grey30") +
geom_label(aes(x=Start[1],y=avgStepsClass,label = round(avgStepsClass, digits = 0),
vjust=-0.5), hjust=-0.01, size=5, color="#199787")
# geom_point(aes(x=Start,y=Hudeliverable), color = "red")
```
### Dieting
```{r}
stepsData$Diet <- as.factor(stepsData$Diet)
steps3 <- stepsData[,c(1,10,17)]
steps3 <- subset(steps3, steps3$Start >= "2020-05-31")
stepsDiet <- subset(steps3, steps3$Diet == "On a diet")
stepsNoDiet <- subset(steps3, steps3$Diet == "Not on a diet")
avgStepsDiet <- mean(stepsDiet$`Steps (count)`)
avgStepsNoDiet <- mean(stepsNoDiet$`Steps (count)`)
ggplot(data = steps3) +
geom_col(data = steps3, aes(x=Start, y=`Steps (count)`, fill = Diet)) +
labs(x="Day", y="Steps per day", fill=" " ) + lims(y=c(0,30000)) +
scale_fill_manual(values = c("grey30", "#AA5FFF")) +
theme_classic() + theme(legend.position = "bottom", text=element_text(size=18)) +
geom_hline(yintercept=avgStepsNoDiet, color = "grey30", size=1.5) +
geom_hline(yintercept=avgStepsDiet, color = "#AA5FFF", size=2) +
geom_label(aes(x=Start[1],y=avgStepsNoDiet,label = round(avgStepsNoDiet, digits = 0),
vjust=1.5), hjust=-0.01, size=5, color="grey30") +
geom_label(aes(x=Start[1],y=avgStepsDiet,label = round(avgStepsDiet, digits = 0),
vjust=-0.5), hjust=-0.01, size=5, color="#AA5FFF")
# geom_point(aes(x=Start,y=Hudeliverable), color = "red")
```
Row {data-height=350}
-----------------------------------------------------------------------
### Text
The charts on this page visualizes the number of steps I have taken in the last 3 to 4 months. I wanted to analyze some of the factors that affect the number of steps that I take each day. I looked at how travelling, HU classes and deliverables, and being on a diet affected my daily steps. From the charts,
1. I took the maximum number of steps on days when I was travelling, followed by
2. Days on which I was on a diet. Being on a diet makes me more active.
3. Lastly, I am least active on days when I either have Harrisburg University classes or have class deliverables due.
Macros
=======================================================================
Rows {data-height=150}
-----------------------------------------------------------------------
#### The charts below show how well I followed the Keto Diet during the first 28 days. As Keto is a high-fat, low-carb diet, my target was to consume a minimum of 55% of my daily calories from fats and a maximum of 10% of my daily calories from carbs. The remaining 35% calories woud come from protein. From these charts, I can see I have done a good job reaching the target for carbs and fats intake for most days. I can see that protien intake has been less than the target for many days. Also, I tend to eat more carbs when I am travelling or eating out.
Row {data-height=650}
-----------------------------------------------------------------------
### Carbs (Target = 10%)
```{r fig.height=7}
macros <- read_xlsx("MacrosData.xlsx")
ggplot(macros, aes(Week, reorder(Day, -Day), fill= Carbs)) +
geom_tile() + labs(y="Day") +
scale_fill_gradient2(low="green", mid="white", high="red", midpoint = 0.1,
breaks = c(0.05, 0.1, 0.15),
labels = c("5%", "10%", "15%")) +
theme_minimal() + theme(text = element_text(size = 18))
```
### Fats (Target = 55%)
```{r fig.height=7}
ggplot(macros, aes(Week, reorder(Day, -Day), fill= Fat)) +
geom_tile() + labs(y="Day") +
scale_fill_gradient2(low="red", mid="white", high="green", midpoint = 0.55,
breaks = c(0.45, 0.55, 0.65, 0.75),
labels = c("45%", "55%", "65%", "75%")) +
theme_minimal() + theme(text = element_text(size = 18))
```
### Protein (Target = 35%)
```{r fig.height=7}
ggplot(macros, aes(Week, reorder(Day, -Day), fill= Protein)) +
geom_tile() + labs(y="Day") +
scale_fill_gradient2(low="red", mid="white", high="green", midpoint = 0.35,
breaks = c(0.15, 0.25, 0.35, 0.45),
labels = c("15%", "25%", "35%", "45%")) +
theme_minimal() + theme(text = element_text(size = 18))
```
Rows {data-height=200}
-----------------------------------------------------------------------
### Mood during first 28 days of Keto
```{r fig.width=12, fig.height=1}
ggplot(macros, aes(Days, Tile, fill= Mood)) +
geom_tile(color = "gray") + theme_classic() + labs(x="Day") +
scale_fill_manual(values = c("#FEA1A1", "#FAFAFA", "#A1FEA4")) +
geom_text(aes(label=Reason), size=3, angle=45) +
theme(text = element_text(size = 10),
axis.title = element_blank(), axis.text.y = element_blank(),
axis.line = element_blank(), axis.ticks = element_blank())
```
Location
=======================================================================
Rows {data-height=50}
-----------------------------------------------------------------------
#### Places I visited during my first month in the United States
Rows {data-height=450}
-----------------------------------------------------------------------
```{r}
raw <- jsonlite::fromJSON("2016_SEPTEMBER.json")
locations <- raw$timelineObjects
locations1 <- cbind.data.frame(locations$activitySegment$startLocation$latitudeE7,
locations$activitySegment$startLocation$longitudeE7,
locations$activitySegment$duration$startTimestampMs)
colnames(locations1) <- c("lat","lon","timestamp")
locations1$lat <- locations1$lat/10000000
locations1$lon <- locations1$lon/10000000
```
### Pittsburgh - Sept 2016
```{r}
leaflet(locations1) %>%
addProviderTiles("Stamen.TonerLite") %>%
setView(-80, 40.44, zoom = 12.6) %>%
addCircles(opacity = 0.5, weight = 20, color = "#FBB73B")
#addHeatmap(data = locations1, lng = ~lon, lat = ~lat)
```
### Cleveland - Sept 2016
```{r}
leaflet(locations1) %>%
addProviderTiles("Stamen.TonerLite") %>%
setView(-81.63, 41.5, zoom = 12.5) %>%
addCircles(opacity = 0.5, weight = 20, color = "#453231")
#addHeatmap(data = locations1, lng = ~lon, lat = ~lat)
```
Row {data-height=500}
-----------------------------------------------------------------------
### Places visited in the US in September 2016
```{r}
leaflet(locations1) %>%
addProviderTiles(providers$CartoDB.Positron) %>%
addMarkers(data = locations1, ~lon, ~lat) #, clusterOptions = markerClusterOptions(), group = "Points") %>%
#addHeatmap(lng = ~lon, lat = ~lat, group = "HeatMap", blur = 20, max = 0.01, radius = 15)
```
