red_amber 0.2.1 → 0.2.3

data/benchmark/drop_nil.yml

@@ -1,11 +1,23 @@
+contexts:
+  - gems:
+      red_amber: 0.1.8
+  - gems:
+      red_amber: 0.2.2
+  - name: HEAD
+    prelude: |
+      $LOAD_PATH.unshift(File.expand_path('lib'))
+      require 'red_amber'
+
 prelude: |
   require 'datasets-arrow'
   require 'red_amber'
 
   penguins = RedAmber::DataFrame.new(Datasets::Penguins.new.to_arrow)
 
-  def drop_nil(penguins)
-    penguins.remove { vectors.map { |v| v.is_nil} }
+  def remove_nil(penguins)
+    penguins.remove { vectors.map(&:is_nil).reduce(&:|) }
   end
 
-benchmark: drop_nil(penguins)
+benchmark:
+  'Remove and reduce': remove_nil(penguins)
+  'remove_nil method': penguins.remove_nil

data/benchmark/group.yml

@@ -0,0 +1,33 @@
+contexts:
+  - name: HEAD
+    prelude: |
+      $LOAD_PATH.unshift(File.expand_path('lib'))
+  - gems:
+      red_amber: 0.2.2
+
+prelude: |
+  require 'red_amber'
+  require 'datasets-arrow'
+
+  ds = Datasets::Rdatasets.new('nycflights13', 'flights')
+  df = RedAmber::DataFrame.new(ds.to_arrow)
+    .assign(:flight) { flight.map(&:to_s) }
+
+  slicer = df[:distance] > 1000
+  distance_km = df[:distance] * 1.852
+
+benchmark:
+  'G01: sum distance by destination': |
+    df.group(:dest).sum(:distance)
+
+  'G02: sum arr_delay by month and day': |
+    df.group(:month, :day).sum(:arr_delay)
+
+  'G03: sum arr_delay, mean distance by flight': |
+    df.group(:flight) { [sum(:arr_delay), mean(:distance)] }
+
+  'G04: mean air_time, distance by flight': |
+    df.group(:flight).mean(:air_time, :distance)
+
+  'G05: sum dep_delay, arr_delay by carrer': |
+    df.group(:carrier).sum(:dep_delay, :arr_delay)

data/benchmark/reshape.yml

@@ -0,0 +1,27 @@
+# --repeat-count 3
+
+contexts:
+  - name: HEAD
+    prelude: |
+      $LOAD_PATH.unshift(File.expand_path('lib'))
+  - gems:
+      red_amber: 0.2.2
+
+prelude: |
+  require 'red_amber'
+  require 'datasets-arrow'
+
+  ds = Datasets::Rdatasets.new('tidyr', 'billboard')
+  df = RedAmber::DataFrame.new(ds.to_arrow)
+  sub_df = df.pick(:track, df.keys.select{ |k| k.start_with? 'wk' })
+  long_df = df.to_long(:artist, :track, :'date.entered', name: :week, value: :rank)
+
+benchmark:
+  'R01: Transpose a DataFrame': |
+    sub_df.transpose(name: :week)
+
+  'R02: Reshape to longer DataFrame': |
+    df.to_long(:artist, :track, :'date.entered', name: :week, value: :rank)
+
+  'R03: Reshape to wider DataFrame': |
+    long_df.to_wide(name: :week, value: :rank)

data/benchmark/{csv_load_penguins.yml → rover/csv_load_penguins.yml}

@@ -2,12 +2,12 @@ prelude: |
   require 'rover'
   require 'red_amber'
 
-  penguins_csv = 'benchmark/cache/penguins.csv'
+  penguins_csv = 'tmp/penguins.csv'
 
   unless File.exist?(penguins_csv)
     require 'datasets-arrow'
-    arrow = Datasets::Penguins.new.to_arrow
-    RedAmber::DataFrame.new(arrow).save(penguins_csv)
+    ds = Datasets::Penguins.new
+    RedAmber::DataFrame.new(ds).save(penguins_csv)
   end
 
 benchmark:

data/benchmark/rover/flights.yml

@@ -0,0 +1,23 @@
+contexts:
+  - gems:
+      red_amber: 0.2.2
+  - name: HEAD
+    prelude: |
+      $LOAD_PATH.unshift(File.expand_path('lib'))
+      require 'red_amber'
+
+prelude: |
+  require 'rover'
+  require 'datasets-arrow'
+  ds = Datasets::Rdatasets.new('nycflights13', 'flights')
+  df = RedAmber::DataFrame.new(ds)
+  rover = Rover::DataFrame.new(df.to_h)
+  group_keys = [:month, :origin]
+  summary_key = :air_time
+
+benchmark:
+  'penguins Group by Rover': |
+    rover.group(group_keys).count
+
+  'penguins Group by RedAmber': |
+    df.group(group_keys).count

data/benchmark/rover/penguins.yml

@@ -0,0 +1,23 @@
+contexts:
+  - gems:
+      red_amber: 0.2.2
+  - name: HEAD
+    prelude: |
+      $LOAD_PATH.unshift(File.expand_path('lib'))
+      require 'red_amber'
+
+prelude: |
+  require 'rover'
+  require 'datasets-arrow'
+  ds = Datasets::Penguins.new
+  df = RedAmber::DataFrame.new(ds)
+  rover = Rover::DataFrame.new(df.to_h)
+  group_keys = [:species, :island]
+  summary_key = :body_mass_g
+
+benchmark:
+  'penguins Group by Rover': |
+    rover.group(group_keys).mean(summary_key)
+
+  'penguins Group by RedAmber': |
+    df.group(group_keys).mean(summary_key)

data/benchmark/rover/planes.yml

@@ -0,0 +1,23 @@
+contexts:
+  - gems:
+      red_amber: 0.2.2
+  - name: HEAD
+    prelude: |
+      $LOAD_PATH.unshift(File.expand_path('lib'))
+      require 'red_amber'
+
+prelude: |
+  require 'rover'
+  require 'datasets-arrow'
+  ds = Datasets::Rdatasets.new('nycflights13', 'planes')
+  df = RedAmber::DataFrame.new(ds)
+  rover = Rover::DataFrame.new(df.to_h)
+  group_keys = [:engines, :engine]
+  summary_key = :seats
+
+benchmark:
+  'penguins Group by Rover': |
+    rover.group(group_keys).mean(summary_key)
+
+  'penguins Group by RedAmber': |
+    df.group(group_keys).mean(summary_key)

data/benchmark/rover/weather.yml

@@ -0,0 +1,23 @@
+contexts:
+  - gems:
+      red_amber: 0.2.2
+  - name: HEAD
+    prelude: |
+      $LOAD_PATH.unshift(File.expand_path('lib'))
+      require 'red_amber'
+
+prelude: |
+  require 'rover'
+  require 'datasets-arrow'
+  ds = Datasets::Rdatasets.new('nycflights13', 'weather')
+  df = RedAmber::DataFrame.new(ds)
+  rover = Rover::DataFrame.new(df.to_h)
+  group_keys = [:month, :origin]
+  summary_key = :temp
+
+benchmark:
+  'penguins Group by Rover': |
+    rover.group(group_keys).mean(summary_key)
+
+  'penguins Group by RedAmber': |
+    df.group(group_keys).mean(summary_key)