lokeshh

require 'benchmark'

n = 10000
df = Daru::DataFrame.new({
  a: Statsample::Shorthand.rnorm(n),
  b: Statsample::Shorthand.rnorm(n),
  c: Statsample::Shorthand.rnorm(n),
  d: Statsample::Shorthand.rnorm(n),
})
df[:y] = df.a * 3 + df.b * 5 + df.c * -2 + df.d * 2

lokeshh

require 'ruby-prof'
n = 100000
dv = Daru::Vector.new 1..n
RubyProf.start
100.times { dv.reject_values nil }
result = RubyProf.stop
printer = RubyProf::GraphHtmlPrinter.new(result)
printer.print(File.open('cache.html', 'w'))

lokeshh

require 'statsample'
require 'benchmark'

n = 10000
df = Daru::DataFrame.new({
  a: Statsample::Shorthand.rnorm(n),
  b: Statsample::Shorthand.rnorm(n),
  c: Statsample::Shorthand.rnorm(n),
  d: Statsample::Shorthand.rnorm(n),
})

lokeshh

Failures:

  1) Daru::DataFrame#summary produces a summary of data frame
     Failure/Error:
       tuple.flatten.each_with_index
            .all? { |tup, i| @levels[i][tup] }
     
     NoMethodError:
       undefined method `flatten' for :report_building_text:Symbol
     # ./lib/daru/index.rb:501:in `include?'

lokeshh

[1] pry(main)> ds = Daru::DataFrame.new({
[1] pry(main)*     a: [1.2, 2.3, 3.1],
[1] pry(main)*     b: [4.20, 5.6, 1.1],
[1] pry(main)* c: [1.1, 2.2, 6.6]})
=> #<Daru::DataFrame(3x3)>
       a   b   c
   0 1.2 4.2 1.1
   1 2.3 5.6 2.2
   2 3.1 1.1 6.6
[2] pry(main)> x = Statsample::Regression.multiple ds, :c

lokeshh

> try = Daru::DataFrame.from_csv 'try.csv'
> Statsample::GLM.compute try, 'y', :logistic
ExceptionForMatrix::ErrNotRegular: Not Regular Matrix
from /home/ubuntu/.rvm/gems/ruby-2.2.3/gems/backports-3.6.8/lib/backports/1.9.2/stdlib/matrix.rb:933:in `block in inverse_from'

lokeshh

[16] pry(main)> k
=> #<Daru::DataFrame(14x4)>
           b  c_no c_yes     y
     0  62.1     1     0     0
     1  74.7     0     1     1
     2  69.7     1     0     1
     3    71     1     0     0
     4  56.9     1     0     1
     5  58.7     1     0     0
     6  63.3     1     0     0

lokeshh

> model = glm(y ~ a + e + c:d + e:d, data=data, family='binomial')
> summary(model)

Call:
glm(formula = y ~ a + e + c:d + e:d, family = "binomial", data = data)

Deviance Residuals: 
    Min       1Q   Median       3Q      Max  
-1.4049  -0.7803  -0.1191   0.9512   1.6154  

lokeshh

y a	b	c	d	e
0	6	62.1	no	male	A
1	18	74.7	yes	female	B
1	16	69.7	no	female	A
0	14	71	no	male	C
1	5	56.9	no	male	B
0	11	58.7	no	female	B
0	8	63.3	no	male	B
1	11	70.4	no	male	A
1	15	70.5	no	male	C

lokeshh

require 'daru'
require 'memory_profiler'
report = MemoryProfiler.report do
  @a = ['a'*1000]*1000
end
report.pretty_print

# Total allocated: 9161 bytes (4 objects)
# Total retained:  9081 bytes (2 objects)