Bagikan melalui

Tutorial: menggunakan R untuk memprediksi harga avocado

  • Artikel

Tutorial ini menyajikan contoh end-to-end dari alur kerja Ilmu Data Synapse di Microsoft Fabric. Ini menggunakan R untuk menganalisis dan memvisualisasikan harga alpukat dalam Amerika Serikat, untuk membangun model pembelajaran mesin yang memprediksi harga alpukat di masa depan.

Tutorial ini mencakup langkah-langkah berikut:

  • Memuat pustaka default
  • Muat data
  • Mengkustomisasi data
  • Menambahkan paket baru ke sesi
  • Menganalisis dan memvisualisasikan data
  • Melatih model

Screenshot of avocados.

Prasyarat

  • Buka atau buat buku catatan. Untuk mempelajari caranya, lihat Cara menggunakan notebook Microsoft Fabric.

  • Atur opsi bahasa ke SparkR (R) untuk mengubah bahasa utama.

  • Lampirkan buku catatan Anda ke lakehouse. Di sisi kiri, pilih Tambahkan untuk menambahkan lakehouse yang ada atau untuk membuat lakehouse.

Memuat pustaka

Gunakan pustaka dari runtime R default:

library(tidyverse)
library(lubridate)
library(hms)

Muat data

Baca harga avocado dari . File CSV, diunduh dari internet:

df <- read.csv('https://synapseaisolutionsa.blob.core.windows.net/public/AvocadoPrice/avocado.csv', header = TRUE)
head(df,5)

Manipulasi data

Pertama, beri kolom nama yang lebih ramah.

# To use lowercase
names(df) <- tolower(names(df))

# To use snake case
avocado <- df %>% 
  rename("av_index" = "x",
         "average_price" = "averageprice",
         "total_volume" = "total.volume",
         "total_bags" = "total.bags",
         "amount_from_small_bags" = "small.bags",
         "amount_from_large_bags" = "large.bags",
         "amount_from_xlarge_bags" = "xlarge.bags")

# Rename codes
avocado2 <- avocado %>% 
  rename("PLU4046" = "x4046",
         "PLU4225" = "x4225",
         "PLU4770" = "x4770")

head(avocado2,5)

Ubah jenis data, hapus kolom yang tidak diinginkan, dan tambahkan konsumsi total:

# Convert data
avocado2$year = as.factor(avocado2$year)
avocado2$date = as.Date(avocado2$date)
avocado2$month  = factor(months(avocado2$date), levels = month.name)
avocado2$average_price =as.numeric(avocado2$average_price)
avocado2$PLU4046 = as.double(avocado2$PLU4046)
avocado2$PLU4225 = as.double(avocado2$PLU4225)
avocado2$PLU4770 = as.double(avocado2$PLU4770)
avocado2$amount_from_small_bags = as.numeric(avocado2$amount_from_small_bags)
avocado2$amount_from_large_bags = as.numeric(avocado2$amount_from_large_bags)
avocado2$amount_from_xlarge_bags = as.numeric(avocado2$amount_from_xlarge_bags)


# Remove unwanted columns
avocado2 <- avocado2 %>% 
  select(-av_index,-total_volume, -total_bags)

# Calculate total consumption 
avocado2 <- avocado2 %>% 
  mutate(total_consumption = PLU4046 + PLU4225 + PLU4770 + amount_from_small_bags + amount_from_large_bags + amount_from_xlarge_bags)

Menginstal paket baru

Gunakan penginstalan paket sebaris untuk menambahkan paket baru ke sesi:

install.packages(c("repr","gridExtra","fpp2"))

Muat pustaka yang diperlukan.

library(tidyverse) 
library(knitr)
library(repr)
library(gridExtra)
library(data.table)

Menganalisis dan memvisualisasikan data

Bandingkan harga avokado konvensional (nonorganik) berdasarkan wilayah:

options(repr.plot.width = 10, repr.plot.height =10)
# filter(mydata, gear %in% c(4,5))
avocado2 %>% 
  filter(region %in% c("PhoenixTucson","Houston","WestTexNewMexico","DallasFtWorth","LosAngeles","Denver","Roanoke","Seattle","Spokane","NewYork")) %>%  
  filter(type == "conventional") %>%           
  select(date, region, average_price) %>% 
  ggplot(aes(x = reorder(region, -average_price, na.rm = T), y = average_price)) +
  geom_jitter(aes(colour = region, alpha = 0.5)) +
  geom_violin(outlier.shape = NA, alpha = 0.5, size = 1) +
  geom_hline(yintercept = 1.5, linetype = 2) +
  geom_hline(yintercept = 1, linetype = 2) +
  annotate("rect", xmin = "LosAngeles", xmax = "PhoenixTucson", ymin = -Inf, ymax = Inf, alpha = 0.2) +
  geom_text(x = "WestTexNewMexico", y = 2.5, label = "My top 5 cities!", hjust = 0.5) +
  stat_summary(fun = "mean") +
  labs(x = "US city",
       y = "Avocado prices", 
       title = "Figure 1. Violin plot of nonorganic avocado prices",
       subtitle = "Visual aids: \n(1) Black dots are average prices of individual avocados by city \n     between January 2015 and March 2018. \n(2) The plot is ordered descendingly.\n(3) The body of the violin becomes fatter when data points increase.") +
  theme_classic() + 
  theme(legend.position = "none", 
        axis.text.x = element_text(angle = 25, vjust = 0.65),
        plot.title = element_text(face = "bold", size = 15)) +
  scale_y_continuous(lim = c(0, 3), breaks = seq(0, 3, 0.5))

Screenshot that shows a graph of nonorganic prices.

Fokus pada wilayah Houston.

library(fpp2)
conv_houston <- avocado2 %>% 
  filter(region == "Houston",
         type == "conventional") %>% 
  group_by(date) %>% 
  summarise(average_price = mean(average_price))
  
# Set up ts   

conv_houston_ts <- ts(conv_houston$average_price,
                 start = c(2015, 1),
                 frequency = 52) 
# Plot

autoplot(conv_houston_ts) +
  labs(title = "Time plot: nonorganic avocado weekly prices in Houston",
       y = "$") +
  geom_point(colour = "brown", shape = 21) +
  geom_path(colour = "brown")

Screenshot of a graph of avocado prices in Houston.

Melatih model pembelajaran mesin

Bangun model prediksi harga untuk area Houston, berdasarkan AutoRegressive Integrated Moving Average (ARIMA):

conv_houston_ts_arima <- auto.arima(conv_houston_ts,
                                    d = 1,
                                    approximation = F,
                                    stepwise = F,
                                    trace = T)

checkresiduals(conv_houston_ts_arima)

Screenshot that shows a graph of residuals.

Tampilkan grafik prakiraan dari model ARIMA Houston:

conv_houston_ts_arima_fc <- forecast(conv_houston_ts_arima, h = 208)

autoplot(conv_houston_ts_arima_fc) + labs(subtitle = "Prediction of weekly prices of nonorganic avocados in Houston",
       y = "$") +
  geom_hline(yintercept = 2.5, linetype = 2, colour = "blue")

Screenshot that shows a graph of forecasts from the ARIMA model.

Konten terkait