galaaz 0.4.8 → 0.4.9

data/blogs/gknit/stats.bib

@@ -0,0 +1,27 @@
+@book{Wilkinson:grammar_of_graphics,
+ author = {Wilkinson, Leland},
+ title = {The Grammar of Graphics (Statistics and Computing)},
+ year = {2005},
+ isbn = {0387245448},
+ publisher = {Springer-Verlag},
+ address = {Berlin, Heidelberg},
+}
+
+@article{Knuth:literate_programming,
+ author = {Knuth, Donald E.},
+ title = {Literate Programming},
+ journal = {Comput. J.},
+ issue_date = {May 1984},
+ volume = {27},
+ number = {2},
+ month = may,
+ year = {1984},
+ issn = {0010-4620},
+ pages = {97--111},
+ numpages = {15},
+ url = {http://dx.doi.org/10.1093/comjnl/27.2.97},
+ doi = {10.1093/comjnl/27.2.97},
+ acmid = {479},
+ publisher = {Oxford University Press},
+ address = {Oxford, UK},
+} 

data/blogs/manual/manual.Rmd

@@ -693,6 +693,93 @@ puts @flights_sm.head.as__data__frame
 puts @flights_sm.head.as__data__frame
 ```
 
+## Summarising data
+
+Function 'summarise' calculates summaries for the data frame. When no 'group_by' is used
+a single value is obtained from the data frame:
+
+```{ruby summarise}
+puts @flights.summarise(delay: E.mean(:dep_delay, na__rm: true)).as__data__frame
+```
+
+When a data frame is groupe with 'group_by' summaries apply to the given group:
+
+```{ruby summarise_group_by}
+by_day = @flights.group_by(:year, :month, :day)
+puts by_day.summarise(delay: :dep_delay.mean(na__rm: true)).head.as__data__frame
+```
+
+Next we put many operations together by pipping them one after the other:
+
+```{ruby pipping}
+delays = @flights.
+           group_by(:dest).
+           summarise(
+             count: E.n,
+             dist: :distance.mean(na__rm: true),
+             delay: :arr_delay.mean(na__rm: true)).
+           filter(:count > 20, :dest != "NHL")
+
+puts delays.as__data__frame
+```
+
+# Using Data Table
+
+```{ruby fread}
+R.library('data.table')
+R.install_and_loads('curl')
+
+input = "https://raw.githubusercontent.com/Rdatatable/data.table/master/vignettes/flights14.csv"
+@flights = R.fread(input)
+puts @flights
+puts @flights.dim
+```
+
+```{ruby data_table}
+
+data_table = R.data__table(
+  ID: R.c("b","b","b","a","a","c"),
+  a: (1..6),
+  b: (7..12),
+  c: (13..18)
+)
+
+puts data_table
+puts data_table.ID
+```
+
+```{ruby subset_i}
+# subset rows in i
+ans = @flights[(:origin.eq "JFK") & (:month.eq 6)]
+puts ans.head
+
+# Get the first two rows from flights.
+
+ans = @flights[(1..2)]
+puts ans
+
+# Sort flights first by column origin in ascending order, and then by dest in descending order:
+
+# ans = @flights[E.order(:origin, -(:dest))]
+# puts ans.head
+
+```
+
+```{ruby select_j}
+# Select column(s) in j
+# select arr_delay column, but return it as a vector.
+
+ans = @flights[:all, :arr_delay]
+puts ans.head
+
+# Select arr_delay column, but return as a data.table instead.
+
+ans = @flights[:all, :arr_delay.list]
+puts ans.head
+
+ans = @flights[:all, E.list(:arr_delay, :dep_delay)]
+```
+
 # Graphics in Galaaz
 
 Creating graphics in Galaaz is quite easy, as it can use all the power of ggplot2.  There are
@@ -731,8 +818,33 @@ the data frame with the necessary data:
 puts @mtcars
 ```
 Now, lets plot the diverging bar plot.  When using gKnit, there is no need to call
-'R.awt' to create a plotting device, since gKnit does take care of it:
-
+'R.awt' to create a plotting device, since gKnit does take care of it. Galaaz 
+provides integration with ggplot. The interested reader should check online for more
+information on ggplot, since it is outside the scope of this manual describing 
+how ggplot works. We give here but a brief description on how this plot is generated.
+
+ggplot implements the 'grammar of graphics'. In this approach, plots are build by
+adding layers to the plot.  On the first layer we describe what we want on the 'x'
+and 'y' axis of the plot.  In this case, we have 'car_name' on the 'x' axis and 
+'mpg\_z' on the 'y' axis. Then the type of graph is specified by adding
+'geom\_bar' (for a bar graph).  We specify that our bars should be filled using 
+'mpg\_type', which is either 'above' or 'bellow' giving then two colours for
+filling. On the next layer we specify the labels for the graph, then we add the
+title and subtitle.  Finally, in a bar chart usually bars go on the vertical direction,
+but in this graph we want the bars to be horizontally layed so we add 'coord\_flip'.
+
+```{ruby diverging_bar, fig.width = 9.1, fig.height = 6.5}
+require 'ggplot'
+
+puts @mtcars.ggplot(E.aes(x: :car_name, y: :mpg_z, label: :mpg_z)) +
+  R.geom_bar(E.aes(fill: :mpg_type), stat: 'identity', width: 0.5) +
+  R.scale_fill_manual(name: 'Mileage',
+                      labels: R.c('Above Average', 'Below Average'),
+                      values: R.c('above': '#00ba38', 'below': '#f8766d')) +
+  R.labs(subtitle: "Normalised mileage from 'mtcars'",
+         title: "Diverging Bars") + 
+  R.coord_flip
+```
 
 
 [TO BE CONTINUED...]

data/blogs/manual/manual.html

@@ -552,22 +552,22 @@ double
 ##  (eval):1:in `exec_ruby'
 ## /home/rbotafogo/desenv/galaaz/lib/util/exec_ruby.rb:141:in `instance_eval'
 ## /home/rbotafogo/desenv/galaaz/lib/util/exec_ruby.rb:141:in `exec_ruby'
-## /home/rbotafogo/desenv/galaaz/lib/gknit/knitr_engine.rb:657:in `block in initialize'
+## /home/rbotafogo/desenv/galaaz/lib/gknit/knitr_engine.rb:650:in `block in initialize'
 ## /home/rbotafogo/desenv/galaaz/lib/R_interface/ruby_callback.rb:77:in `call'
 ## /home/rbotafogo/desenv/galaaz/lib/R_interface/ruby_callback.rb:77:in `callback'
 ## (eval):3:in `function(...) {\n          rb_method(...)'
 ## unknown.r:1:in `in_dir'
 ## unknown.r:1:in `block_exec:BLOCK0'
-## /home/rbotafogo/lib/graalvm-ce-1.0.0-rc15/jre/languages/R/library/knitr/R/block.R:102:in `block_exec'
-## /home/rbotafogo/lib/graalvm-ce-1.0.0-rc15/jre/languages/R/library/knitr/R/block.R:92:in `call_block'
-## /home/rbotafogo/lib/graalvm-ce-1.0.0-rc15/jre/languages/R/library/knitr/R/block.R:6:in `process_group.block'
-## /home/rbotafogo/lib/graalvm-ce-1.0.0-rc15/jre/languages/R/library/knitr/R/block.R:3:in `<no source>'
+## /home/rbotafogo/lib/graalvm-ce-1.0.0-rc16/jre/languages/R/library/knitr/R/block.R:102:in `block_exec'
+## /home/rbotafogo/lib/graalvm-ce-1.0.0-rc16/jre/languages/R/library/knitr/R/block.R:92:in `call_block'
+## /home/rbotafogo/lib/graalvm-ce-1.0.0-rc16/jre/languages/R/library/knitr/R/block.R:6:in `process_group.block'
+## /home/rbotafogo/lib/graalvm-ce-1.0.0-rc16/jre/languages/R/library/knitr/R/block.R:3:in `<no source>'
 ## unknown.r:1:in `withCallingHandlers'
 ## unknown.r:1:in `process_file'
 ## unknown.r:1:in `<no source>:BLOCK1'
-## /home/rbotafogo/lib/graalvm-ce-1.0.0-rc15/jre/languages/R/library/knitr/R/output.R:129:in `<no source>'
+## /home/rbotafogo/lib/graalvm-ce-1.0.0-rc16/jre/languages/R/library/knitr/R/output.R:129:in `<no source>'
 ## unknown.r:1:in `<no source>:BLOCK1'
-## /home/rbotafogo/lib/graalvm-ce-1.0.0-rc15/jre/languages/R/library/rmarkdown/R/render.R:162:in `<no source>'
+## /home/rbotafogo/lib/graalvm-ce-1.0.0-rc16/jre/languages/R/library/rmarkdown/R/render.R:162:in `<no source>'
 ## <REPL>:5:in `<repl wrapper>'
 ## &lt;REPL&gt;:1</code></pre>
 <p>Here is a vector with logical values</p>
@@ -2385,6 +2385,237 @@ puts @flights_sm.head.as__data__frame</code></pre>
 ## 5 2013     1   1        -6       -25      762      116   19 394.1379
 ## 6 2013     1   1        -4        12      719      150  -16 287.6000</code></pre>
 </div>
+<div id="summarising-data" class="section level2">
+<h2>Summarising data</h2>
+<p>Function ‘summarise’ calculates summaries for the data frame. When no ‘group_by’ is used a single value is obtained from the data frame:</p>
+<pre class="ruby"><code>puts @flights.summarise(delay: E.mean(:dep_delay, na__rm: true)).as__data__frame</code></pre>
+<pre><code>##      delay
+## 1 12.63907</code></pre>
+<p>When a data frame is groupe with ‘group_by’ summaries apply to the given group:</p>
+<pre class="ruby"><code>by_day = @flights.group_by(:year, :month, :day)
+puts by_day.summarise(delay: :dep_delay.mean(na__rm: true)).head.as__data__frame</code></pre>
+<pre><code>##   year month day     delay
+## 1 2013     1   1 11.548926
+## 2 2013     1   2 13.858824
+## 3 2013     1   3 10.987832
+## 4 2013     1   4  8.951595
+## 5 2013     1   5  5.732218
+## 6 2013     1   6  7.148014</code></pre>
+<p>Next we put many operations together by pipping them one after the other:</p>
+<pre class="ruby"><code>delays = @flights.
+           group_by(:dest).
+           summarise(
+             count: E.n,
+             dist: :distance.mean(na__rm: true),
+             delay: :arr_delay.mean(na__rm: true)).
+           filter(:count &gt; 20, :dest != &quot;NHL&quot;)
+
+puts delays.as__data__frame</code></pre>
+<pre><code>##    dest count       dist       delay
+## 1   ABQ   254 1826.00000  4.38188976
+## 2   ACK   265  199.00000  4.85227273
+## 3   ALB   439  143.00000 14.39712919
+## 4   ATL 17215  757.10822 11.30011285
+## 5   AUS  2439 1514.25297  6.01990875
+## 6   AVL   275  583.58182  8.00383142
+## 7   BDL   443  116.00000  7.04854369
+## 8   BGR   375  378.00000  8.02793296
+## 9   BHM   297  865.99663 16.87732342
+## 10  BNA  6333  758.21348 11.81245891
+## 11  BOS 15508  190.63696  2.91439222
+## 12  BQN   896 1578.98326  8.24549550
+## 13  BTV  2589  265.09154  8.95099602
+## 14  BUF  4681  296.80837  8.94595186
+## 15  BUR   371 2465.00000  8.17567568
+## 16  BWI  1781  179.41830 10.72673385
+## 17  BZN    36 1882.00000  7.60000000
+## 18  CAE   116  603.55172 41.76415094
+## 19  CAK   864  397.00000 19.69833729
+## 20  CHO    52  305.00000  9.50000000
+## 21  CHS  2884  632.91678 10.59296847
+## 22  CLE  4573  414.17428  9.18161129
+## 23  CLT 14064  538.02730  7.36031885
+## 24  CMH  3524  476.55505 10.60132291
+## 25  CRW   138  444.00000 14.67164179
+## 26  CVG  3941  575.15986 15.36456376
+## 27  DAY  1525  537.10230 12.68048606
+## 28  DCA  9705  211.00618  9.06695204
+## 29  DEN  7266 1614.67836  8.60650021
+## 30  DFW  8738 1383.04303  0.32212685
+## 31  DSM   569 1020.88752 19.00573614
+## 32  DTW  9384  498.12852  5.42996346
+## 33  EGE   213 1735.70892  6.30434783
+## 34  FLL 12055 1070.06877  8.08212154
+## 35  GRR   765  605.78170 18.18956044
+## 36  GSO  1606  449.84184 14.11260054
+## 37  GSP   849  595.95995 15.93544304
+## 38  HNL   707 4972.67468 -1.36519258
+## 39  HOU  2115 1420.15508  7.17618819
+## 40  IAD  5700  224.84684 13.86420212
+## 41  IAH  7198 1407.20672  4.24079040
+## 42  ILM   110  500.00000  4.63551402
+## 43  IND  2077  652.26288  9.94043412
+## 44  JAC    25 1875.60000 28.09523810
+## 45  JAX  2720  824.67610 11.84483416
+## 46  LAS  5997 2240.96148  0.25772849
+## 47  LAX 16174 2468.62236  0.54711094
+## 48  LGB   668 2465.00000 -0.06202723
+## 49  MCI  2008 1097.69522 14.51405836
+## 50  MCO 14082  943.11057  5.45464309
+## 51  MDW  4113  718.04595 12.36422360
+## 52  MEM  1789  954.20123 10.64531435
+## 53  MHT  1009  207.02973 14.78755365
+## 54  MIA 11728 1091.55244  0.29905978
+## 55  MKE  2802  733.38151 14.16722038
+## 56  MSN   572  803.95455 20.19604317
+## 57  MSP  7185 1017.40167  7.27016886
+## 58  MSY  3799 1177.70571  6.49017497
+## 59  MVY   221  173.00000 -0.28571429
+## 60  MYR    59  550.66102  4.60344828
+## 61  OAK   312 2576.00000  3.07766990
+## 62  OKC   346 1325.00000 30.61904762
+## 63  OMA   849 1135.56655 14.69889841
+## 64  ORD 17283  729.00081  5.87661475
+## 65  ORF  1536  288.52344 10.94909344
+## 66  PBI  6554 1028.83811  8.56297210
+## 67  PDX  1354 2445.56573  5.14157973
+## 68  PHL  1632   94.32353 10.12719014
+## 69  PHX  4656 2141.30326  2.09704733
+## 70  PIT  2875  334.06122  7.68099053
+## 71  PSE   365 1617.00000  7.87150838
+## 72  PVD   376  160.00000 16.23463687
+## 73  PWM  2352  276.12840 11.66040210
+## 74  RDU  8163  426.75769 10.05238095
+## 75  RIC  2454  281.40465 20.11125320
+## 76  ROC  2416  259.25083 11.56064461
+## 77  RSW  3537 1072.85327  3.23814963
+## 78  SAN  2737 2437.29923  3.13916574
+## 79  SAT   686 1578.34111  6.94537178
+## 80  SAV   804  709.18408 15.12950601
+## 81  SDF  1157  645.98358 12.66938406
+## 82  SEA  3923 2412.66531 -1.09909910
+## 83  SFO 13331 2577.92356  2.67289152
+## 84  SJC   329 2569.00000  3.44817073
+## 85  SJU  5819 1599.83365  2.52052659
+## 86  SLC  2467 1986.98662  0.17625459
+## 87  SMF   284 2521.00000 12.10992908
+## 88  SNA   825 2434.00000 -7.86822660
+## 89  SRQ  1211 1044.65153  3.08243131
+## 90  STL  4339  878.72321 11.07846451
+## 91  STT   522 1626.98276 -3.83590734
+## 92  SYR  1761  205.92164  8.90392501
+## 93  TPA  7466 1003.93557  7.40852503
+## 94  TUL   315 1215.00000 33.65986395
+## 95  TVC   101  652.38614 12.96842105
+## 96  TYS   631  638.80983 24.06920415
+## 97  XNA  1036 1142.50579  7.46572581</code></pre>
+</div>
+</div>
+<div id="using-data-table" class="section level1">
+<h1>Using Data Table</h1>
+<pre class="ruby"><code>R.library('data.table')
+R.install_and_loads('curl')
+
+input = &quot;https://raw.githubusercontent.com/Rdatatable/data.table/master/vignettes/flights14.csv&quot;
+@flights = R.fread(input)
+puts @flights
+puts @flights.dim</code></pre>
+<pre><code>##         year month day dep_delay arr_delay carrier origin dest air_time
+##      1: 2014     1   1        14        13      AA    JFK  LAX      359
+##      2: 2014     1   1        -3        13      AA    JFK  LAX      363
+##      3: 2014     1   1         2         9      AA    JFK  LAX      351
+##      4: 2014     1   1        -8       -26      AA    LGA  PBI      157
+##      5: 2014     1   1         2         1      AA    JFK  LAX      350
+##     ---                                                                
+## 253312: 2014    10  31         1       -30      UA    LGA  IAH      201
+## 253313: 2014    10  31        -5       -14      UA    EWR  IAH      189
+## 253314: 2014    10  31        -8        16      MQ    LGA  RDU       83
+## 253315: 2014    10  31        -4        15      MQ    LGA  DTW       75
+## 253316: 2014    10  31        -5         1      MQ    LGA  SDF      110
+##         distance hour
+##      1:     2475    9
+##      2:     2475   11
+##      3:     2475   19
+##      4:     1035    7
+##      5:     2475   13
+##     ---              
+## 253312:     1416   14
+## 253313:     1400    8
+## 253314:      431   11
+## 253315:      502   11
+## 253316:      659    8
+## [1] 253316     11</code></pre>
+<pre class="ruby"><code>
+data_table = R.data__table(
+  ID: R.c(&quot;b&quot;,&quot;b&quot;,&quot;b&quot;,&quot;a&quot;,&quot;a&quot;,&quot;c&quot;),
+  a: (1..6),
+  b: (7..12),
+  c: (13..18)
+)
+
+puts data_table
+puts data_table.ID</code></pre>
+<pre><code>##    ID a  b  c
+## 1:  b 1  7 13
+## 2:  b 2  8 14
+## 3:  b 3  9 15
+## 4:  a 4 10 16
+## 5:  a 5 11 17
+## 6:  c 6 12 18
+## [1] &quot;b&quot; &quot;b&quot; &quot;b&quot; &quot;a&quot; &quot;a&quot; &quot;c&quot;</code></pre>
+<pre class="ruby"><code># subset rows in i
+ans = @flights[(:origin.eq &quot;JFK&quot;) &amp; (:month.eq 6)]
+puts ans.head
+
+# Get the first two rows from flights.
+
+ans = @flights[(1..2)]
+puts ans
+
+# Sort flights first by column origin in ascending order, and then by dest in descending order:
+
+# ans = @flights[E.order(:origin, -(:dest))]
+# puts ans.head</code></pre>
+<pre><code>##    year month day dep_delay arr_delay carrier origin dest air_time
+## 1: 2014     6   1        -9        -5      AA    JFK  LAX      324
+## 2: 2014     6   1       -10       -13      AA    JFK  LAX      329
+## 3: 2014     6   1        18        -1      AA    JFK  LAX      326
+## 4: 2014     6   1        -6       -16      AA    JFK  LAX      320
+## 5: 2014     6   1        -4       -45      AA    JFK  LAX      326
+## 6: 2014     6   1        -6       -23      AA    JFK  LAX      329
+##    distance hour
+## 1:     2475    8
+## 2:     2475   12
+## 3:     2475    7
+## 4:     2475   10
+## 5:     2475   18
+## 6:     2475   14
+##    year month day dep_delay arr_delay carrier origin dest air_time
+## 1: 2014     1   1        14        13      AA    JFK  LAX      359
+## 2: 2014     1   1        -3        13      AA    JFK  LAX      363
+##    distance hour
+## 1:     2475    9
+## 2:     2475   11</code></pre>
+<pre class="ruby"><code># Select column(s) in j
+# select arr_delay column, but return it as a vector.
+
+ans = @flights[:all, :arr_delay]
+puts ans.head
+
+# Select arr_delay column, but return as a data.table instead.
+
+ans = @flights[:all, :arr_delay.list]
+puts ans.head
+
+ans = @flights[:all, E.list(:arr_delay, :dep_delay)]</code></pre>
+<pre><code>## [1]  13  13   9 -26   1   0
+##    arr_delay
+## 1:        13
+## 2:        13
+## 3:         9
+## 4:       -26
+## 5:         1
+## 6:         0</code></pre>
 </div>
 <div id="graphics-in-galaaz" class="section level1">
 <h1>Graphics in Galaaz</h1>
@@ -2482,7 +2713,19 @@ puts @mtcars</code></pre>
 ## Lotus Europa               Lotus Europa  1.71    above
 ## Fiat 128                       Fiat 128  2.04    above
 ## Toyota Corolla           Toyota Corolla  2.29    above</code></pre>
-<p>Now, lets plot the diverging bar plot. When using gKnit, there is no need to call ‘R.awt’ to create a plotting device, since gKnit does take care of it:</p>
+<p>Now, lets plot the diverging bar plot. When using gKnit, there is no need to call ‘R.awt’ to create a plotting device, since gKnit does take care of it. Galaaz provides integration with ggplot. The interested reader should check online for more information on ggplot, since it is outside the scope of this manual describing how ggplot works. We give here but a brief description on how this plot is generated.</p>
+<p>ggplot implements the ‘grammar of graphics’. In this approach, plots are build by adding layers to the plot. On the first layer we describe what we want on the ‘x’ and ‘y’ axis of the plot. In this case, we have ‘car_name’ on the ‘x’ axis and ‘mpg_z’ on the ‘y’ axis. Then the type of graph is specified by adding ‘geom_bar’ (for a bar graph). We specify that our bars should be filled using ‘mpg_type’, which is either ‘above’ or ‘bellow’ giving then two colours for filling. On the next layer we specify the labels for the graph, then we add the title and subtitle. Finally, in a bar chart usually bars go on the vertical direction, but in this graph we want the bars to be horizontally layed so we add ‘coord_flip’.</p>
+<pre class="ruby"><code>require 'ggplot'
+
+puts @mtcars.ggplot(E.aes(x: :car_name, y: :mpg_z, label: :mpg_z)) +
+  R.geom_bar(E.aes(fill: :mpg_type), stat: 'identity', width: 0.5) +
+  R.scale_fill_manual(name: 'Mileage',
+                      labels: R.c('Above Average', 'Below Average'),
+                      values: R.c('above': '#00ba38', 'below': '#f8766d')) +
+  R.labs(subtitle: &quot;Normalised mileage from 'mtcars'&quot;,
+         title: &quot;Diverging Bars&quot;) + 
+  R.coord_flip</code></pre>
+<p><img src="data:image/png;base64,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" /><!-- --></p>
 <p>[TO BE CONTINUED…]</p>
 </div>
 <div id="contributing" class="section level1">

data/blogs/manual/manual.md

@@ -221,22 +221,22 @@ vec = R.c(1, hello, 5)
 ##  (eval):1:in `exec_ruby'
 ## /home/rbotafogo/desenv/galaaz/lib/util/exec_ruby.rb:141:in `instance_eval'
 ## /home/rbotafogo/desenv/galaaz/lib/util/exec_ruby.rb:141:in `exec_ruby'
-## /home/rbotafogo/desenv/galaaz/lib/gknit/knitr_engine.rb:657:in `block in initialize'
+## /home/rbotafogo/desenv/galaaz/lib/gknit/knitr_engine.rb:650:in `block in initialize'
 ## /home/rbotafogo/desenv/galaaz/lib/R_interface/ruby_callback.rb:77:in `call'
 ## /home/rbotafogo/desenv/galaaz/lib/R_interface/ruby_callback.rb:77:in `callback'
 ## (eval):3:in `function(...) {\n          rb_method(...)'
 ## unknown.r:1:in `in_dir'
 ## unknown.r:1:in `block_exec:BLOCK0'
-## /home/rbotafogo/lib/graalvm-ce-1.0.0-rc15/jre/languages/R/library/knitr/R/block.R:102:in `block_exec'
-## /home/rbotafogo/lib/graalvm-ce-1.0.0-rc15/jre/languages/R/library/knitr/R/block.R:92:in `call_block'
-## /home/rbotafogo/lib/graalvm-ce-1.0.0-rc15/jre/languages/R/library/knitr/R/block.R:6:in `process_group.block'
-## /home/rbotafogo/lib/graalvm-ce-1.0.0-rc15/jre/languages/R/library/knitr/R/block.R:3:in `<no source>'
+## /home/rbotafogo/lib/graalvm-ce-1.0.0-rc16/jre/languages/R/library/knitr/R/block.R:102:in `block_exec'
+## /home/rbotafogo/lib/graalvm-ce-1.0.0-rc16/jre/languages/R/library/knitr/R/block.R:92:in `call_block'
+## /home/rbotafogo/lib/graalvm-ce-1.0.0-rc16/jre/languages/R/library/knitr/R/block.R:6:in `process_group.block'
+## /home/rbotafogo/lib/graalvm-ce-1.0.0-rc16/jre/languages/R/library/knitr/R/block.R:3:in `<no source>'
 ## unknown.r:1:in `withCallingHandlers'
 ## unknown.r:1:in `process_file'
 ## unknown.r:1:in `<no source>:BLOCK1'
-## /home/rbotafogo/lib/graalvm-ce-1.0.0-rc15/jre/languages/R/library/knitr/R/output.R:129:in `<no source>'
+## /home/rbotafogo/lib/graalvm-ce-1.0.0-rc16/jre/languages/R/library/knitr/R/output.R:129:in `<no source>'
 ## unknown.r:1:in `<no source>:BLOCK1'
-## /home/rbotafogo/lib/graalvm-ce-1.0.0-rc15/jre/languages/R/library/rmarkdown/R/render.R:162:in `<no source>'
+## /home/rbotafogo/lib/graalvm-ce-1.0.0-rc16/jre/languages/R/library/rmarkdown/R/render.R:162:in `<no source>'
 ## <REPL>:5:in `<repl wrapper>'
 ## <REPL>:1
 ```
@@ -1764,6 +1764,288 @@ puts @flights_sm.head.as__data__frame
 ## 6 2013     1   1        -4        12      719      150  -16 287.6000
 ```
 
+## Summarising data
+
+Function 'summarise' calculates summaries for the data frame. When no 'group_by' is used
+a single value is obtained from the data frame:
+
+
+```ruby
+puts @flights.summarise(delay: E.mean(:dep_delay, na__rm: true)).as__data__frame
+```
+
+```
+##      delay
+## 1 12.63907
+```
+
+When a data frame is groupe with 'group_by' summaries apply to the given group:
+
+
+```ruby
+by_day = @flights.group_by(:year, :month, :day)
+puts by_day.summarise(delay: :dep_delay.mean(na__rm: true)).head.as__data__frame
+```
+
+```
+##   year month day     delay
+## 1 2013     1   1 11.548926
+## 2 2013     1   2 13.858824
+## 3 2013     1   3 10.987832
+## 4 2013     1   4  8.951595
+## 5 2013     1   5  5.732218
+## 6 2013     1   6  7.148014
+```
+
+Next we put many operations together by pipping them one after the other:
+
+
+```ruby
+delays = @flights.
+           group_by(:dest).
+           summarise(
+             count: E.n,
+             dist: :distance.mean(na__rm: true),
+             delay: :arr_delay.mean(na__rm: true)).
+           filter(:count > 20, :dest != "NHL")
+
+puts delays.as__data__frame
+```
+
+```
+##    dest count       dist       delay
+## 1   ABQ   254 1826.00000  4.38188976
+## 2   ACK   265  199.00000  4.85227273
+## 3   ALB   439  143.00000 14.39712919
+## 4   ATL 17215  757.10822 11.30011285
+## 5   AUS  2439 1514.25297  6.01990875
+## 6   AVL   275  583.58182  8.00383142
+## 7   BDL   443  116.00000  7.04854369
+## 8   BGR   375  378.00000  8.02793296
+## 9   BHM   297  865.99663 16.87732342
+## 10  BNA  6333  758.21348 11.81245891
+## 11  BOS 15508  190.63696  2.91439222
+## 12  BQN   896 1578.98326  8.24549550
+## 13  BTV  2589  265.09154  8.95099602
+## 14  BUF  4681  296.80837  8.94595186
+## 15  BUR   371 2465.00000  8.17567568
+## 16  BWI  1781  179.41830 10.72673385
+## 17  BZN    36 1882.00000  7.60000000
+## 18  CAE   116  603.55172 41.76415094
+## 19  CAK   864  397.00000 19.69833729
+## 20  CHO    52  305.00000  9.50000000
+## 21  CHS  2884  632.91678 10.59296847
+## 22  CLE  4573  414.17428  9.18161129
+## 23  CLT 14064  538.02730  7.36031885
+## 24  CMH  3524  476.55505 10.60132291
+## 25  CRW   138  444.00000 14.67164179
+## 26  CVG  3941  575.15986 15.36456376
+## 27  DAY  1525  537.10230 12.68048606
+## 28  DCA  9705  211.00618  9.06695204
+## 29  DEN  7266 1614.67836  8.60650021
+## 30  DFW  8738 1383.04303  0.32212685
+## 31  DSM   569 1020.88752 19.00573614
+## 32  DTW  9384  498.12852  5.42996346
+## 33  EGE   213 1735.70892  6.30434783
+## 34  FLL 12055 1070.06877  8.08212154
+## 35  GRR   765  605.78170 18.18956044
+## 36  GSO  1606  449.84184 14.11260054
+## 37  GSP   849  595.95995 15.93544304
+## 38  HNL   707 4972.67468 -1.36519258
+## 39  HOU  2115 1420.15508  7.17618819
+## 40  IAD  5700  224.84684 13.86420212
+## 41  IAH  7198 1407.20672  4.24079040
+## 42  ILM   110  500.00000  4.63551402
+## 43  IND  2077  652.26288  9.94043412
+## 44  JAC    25 1875.60000 28.09523810
+## 45  JAX  2720  824.67610 11.84483416
+## 46  LAS  5997 2240.96148  0.25772849
+## 47  LAX 16174 2468.62236  0.54711094
+## 48  LGB   668 2465.00000 -0.06202723
+## 49  MCI  2008 1097.69522 14.51405836
+## 50  MCO 14082  943.11057  5.45464309
+## 51  MDW  4113  718.04595 12.36422360
+## 52  MEM  1789  954.20123 10.64531435
+## 53  MHT  1009  207.02973 14.78755365
+## 54  MIA 11728 1091.55244  0.29905978
+## 55  MKE  2802  733.38151 14.16722038
+## 56  MSN   572  803.95455 20.19604317
+## 57  MSP  7185 1017.40167  7.27016886
+## 58  MSY  3799 1177.70571  6.49017497
+## 59  MVY   221  173.00000 -0.28571429
+## 60  MYR    59  550.66102  4.60344828
+## 61  OAK   312 2576.00000  3.07766990
+## 62  OKC   346 1325.00000 30.61904762
+## 63  OMA   849 1135.56655 14.69889841
+## 64  ORD 17283  729.00081  5.87661475
+## 65  ORF  1536  288.52344 10.94909344
+## 66  PBI  6554 1028.83811  8.56297210
+## 67  PDX  1354 2445.56573  5.14157973
+## 68  PHL  1632   94.32353 10.12719014
+## 69  PHX  4656 2141.30326  2.09704733
+## 70  PIT  2875  334.06122  7.68099053
+## 71  PSE   365 1617.00000  7.87150838
+## 72  PVD   376  160.00000 16.23463687
+## 73  PWM  2352  276.12840 11.66040210
+## 74  RDU  8163  426.75769 10.05238095
+## 75  RIC  2454  281.40465 20.11125320
+## 76  ROC  2416  259.25083 11.56064461
+## 77  RSW  3537 1072.85327  3.23814963
+## 78  SAN  2737 2437.29923  3.13916574
+## 79  SAT   686 1578.34111  6.94537178
+## 80  SAV   804  709.18408 15.12950601
+## 81  SDF  1157  645.98358 12.66938406
+## 82  SEA  3923 2412.66531 -1.09909910
+## 83  SFO 13331 2577.92356  2.67289152
+## 84  SJC   329 2569.00000  3.44817073
+## 85  SJU  5819 1599.83365  2.52052659
+## 86  SLC  2467 1986.98662  0.17625459
+## 87  SMF   284 2521.00000 12.10992908
+## 88  SNA   825 2434.00000 -7.86822660
+## 89  SRQ  1211 1044.65153  3.08243131
+## 90  STL  4339  878.72321 11.07846451
+## 91  STT   522 1626.98276 -3.83590734
+## 92  SYR  1761  205.92164  8.90392501
+## 93  TPA  7466 1003.93557  7.40852503
+## 94  TUL   315 1215.00000 33.65986395
+## 95  TVC   101  652.38614 12.96842105
+## 96  TYS   631  638.80983 24.06920415
+## 97  XNA  1036 1142.50579  7.46572581
+```
+
+# Using Data Table
+
+
+```ruby
+R.library('data.table')
+R.install_and_loads('curl')
+
+input = "https://raw.githubusercontent.com/Rdatatable/data.table/master/vignettes/flights14.csv"
+@flights = R.fread(input)
+puts @flights
+puts @flights.dim
+```
+
+```
+##         year month day dep_delay arr_delay carrier origin dest air_time
+##      1: 2014     1   1        14        13      AA    JFK  LAX      359
+##      2: 2014     1   1        -3        13      AA    JFK  LAX      363
+##      3: 2014     1   1         2         9      AA    JFK  LAX      351
+##      4: 2014     1   1        -8       -26      AA    LGA  PBI      157
+##      5: 2014     1   1         2         1      AA    JFK  LAX      350
+##     ---                                                                
+## 253312: 2014    10  31         1       -30      UA    LGA  IAH      201
+## 253313: 2014    10  31        -5       -14      UA    EWR  IAH      189
+## 253314: 2014    10  31        -8        16      MQ    LGA  RDU       83
+## 253315: 2014    10  31        -4        15      MQ    LGA  DTW       75
+## 253316: 2014    10  31        -5         1      MQ    LGA  SDF      110
+##         distance hour
+##      1:     2475    9
+##      2:     2475   11
+##      3:     2475   19
+##      4:     1035    7
+##      5:     2475   13
+##     ---              
+## 253312:     1416   14
+## 253313:     1400    8
+## 253314:      431   11
+## 253315:      502   11
+## 253316:      659    8
+## [1] 253316     11
+```
+
+
+```ruby
+
+data_table = R.data__table(
+  ID: R.c("b","b","b","a","a","c"),
+  a: (1..6),
+  b: (7..12),
+  c: (13..18)
+)
+
+puts data_table
+puts data_table.ID
+```
+
+```
+##    ID a  b  c
+## 1:  b 1  7 13
+## 2:  b 2  8 14
+## 3:  b 3  9 15
+## 4:  a 4 10 16
+## 5:  a 5 11 17
+## 6:  c 6 12 18
+## [1] "b" "b" "b" "a" "a" "c"
+```
+
+
+```ruby
+# subset rows in i
+ans = @flights[(:origin.eq "JFK") & (:month.eq 6)]
+puts ans.head
+
+# Get the first two rows from flights.
+
+ans = @flights[(1..2)]
+puts ans
+
+# Sort flights first by column origin in ascending order, and then by dest in descending order:
+
+# ans = @flights[E.order(:origin, -(:dest))]
+# puts ans.head
+```
+
+```
+##    year month day dep_delay arr_delay carrier origin dest air_time
+## 1: 2014     6   1        -9        -5      AA    JFK  LAX      324
+## 2: 2014     6   1       -10       -13      AA    JFK  LAX      329
+## 3: 2014     6   1        18        -1      AA    JFK  LAX      326
+## 4: 2014     6   1        -6       -16      AA    JFK  LAX      320
+## 5: 2014     6   1        -4       -45      AA    JFK  LAX      326
+## 6: 2014     6   1        -6       -23      AA    JFK  LAX      329
+##    distance hour
+## 1:     2475    8
+## 2:     2475   12
+## 3:     2475    7
+## 4:     2475   10
+## 5:     2475   18
+## 6:     2475   14
+##    year month day dep_delay arr_delay carrier origin dest air_time
+## 1: 2014     1   1        14        13      AA    JFK  LAX      359
+## 2: 2014     1   1        -3        13      AA    JFK  LAX      363
+##    distance hour
+## 1:     2475    9
+## 2:     2475   11
+```
+
+
+```ruby
+# Select column(s) in j
+# select arr_delay column, but return it as a vector.
+
+ans = @flights[:all, :arr_delay]
+puts ans.head
+
+# Select arr_delay column, but return as a data.table instead.
+
+ans = @flights[:all, :arr_delay.list]
+puts ans.head
+
+ans = @flights[:all, E.list(:arr_delay, :dep_delay)]
+```
+
+```
+## [1]  13  13   9 -26   1   0
+##    arr_delay
+## 1:        13
+## 2:        13
+## 3:         9
+## 4:       -26
+## 5:         1
+## 6:         0
+```
+
 # Graphics in Galaaz
 
 Creating graphics in Galaaz is quite easy, as it can use all the power of ggplot2.  There are
@@ -1872,8 +2154,37 @@ puts @mtcars
 ## Toyota Corolla           Toyota Corolla  2.29    above
 ```
 Now, lets plot the diverging bar plot.  When using gKnit, there is no need to call
-'R.awt' to create a plotting device, since gKnit does take care of it:
+'R.awt' to create a plotting device, since gKnit does take care of it. Galaaz 
+provides integration with ggplot. The interested reader should check online for more
+information on ggplot, since it is outside the scope of this manual describing 
+how ggplot works. We give here but a brief description on how this plot is generated.
+
+ggplot implements the 'grammar of graphics'. In this approach, plots are build by
+adding layers to the plot.  On the first layer we describe what we want on the 'x'
+and 'y' axis of the plot.  In this case, we have 'car_name' on the 'x' axis and 
+'mpg\_z' on the 'y' axis. Then the type of graph is specified by adding
+'geom\_bar' (for a bar graph).  We specify that our bars should be filled using 
+'mpg\_type', which is either 'above' or 'bellow' giving then two colours for
+filling. On the next layer we specify the labels for the graph, then we add the
+title and subtitle.  Finally, in a bar chart usually bars go on the vertical direction,
+but in this graph we want the bars to be horizontally layed so we add 'coord\_flip'.
+
+
+```ruby
+require 'ggplot'
+
+puts @mtcars.ggplot(E.aes(x: :car_name, y: :mpg_z, label: :mpg_z)) +
+  R.geom_bar(E.aes(fill: :mpg_type), stat: 'identity', width: 0.5) +
+  R.scale_fill_manual(name: 'Mileage',
+                      labels: R.c('Above Average', 'Below Average'),
+                      values: R.c('above': '#00ba38', 'below': '#f8766d')) +
+  R.labs(subtitle: "Normalised mileage from 'mtcars'",
+         title: "Diverging Bars") + 
+  R.coord_flip
+```
+
 
+![](/home/rbotafogo/desenv/galaaz/blogs/manual/manual_files/figure-html/diverging_bar.png)<!-- -->
 
 
 [TO BE CONTINUED...]