saptarshiguha

REATE OR REPLACE FUNCTION  analysis.sg_histogram_aggregate(r  ARRAY<STRUCT<v STRING>>) RETURNS Array<STRUCT<k STRING, v INT64>>
LANGUAGE js AS \"\"\"
// called as sg_histogram_aggregate(ARRAY_AGG(Struct( JSON_EXTRACT(A HIstogram,'$.values') as v)))
var arrayLength = r.length;
if( arrayLength == 0) {
   return( [] );
}
var d = 0;
var accum = {  };
for (var i = 0; i < arrayLength; i++) {

saptarshiguha

output
html_document keep_md true

Approaches to Estimating Samples Sizes and Power

Background

As data scientist at Mozilla one of your responsibilities will be to analyze

saptarshiguha

output
html_document keep_md true

Approaches to Estimating Samples Sizes and Power

Background

As data scientist at Mozilla one of your responsibilities will be to analyze

saptarshiguha

w=spark.sql("""
select
submission_date_s3,
client_id as cid,
sum(coalesce(scalar_parent_browser_engagement_total_uri_count,0)) as turi,
case when sum(coalesce(scalar_parent_browser_engagement_total_uri_count,0)) >=5 then 1 else 0 end as adau,
cast(sum(coalesce(scalar_parent_browser_engagement_total_uri_count,0))/(sum(active_ticks*5.0/3600)) as float) as turihr
from main_summary
where submission_date_s3>='20180701' and submission_date_s3<='20180707'

saptarshiguha

(custom-set-variables '(package-archives
  (quote
   (
    ;; ("marmalade" . "http://marmalade-repo.org/packages/")
    ("melpa" . "http://melpa.org/packages/")
    ("gnu" . "http://elpa.gnu.org/packages/")))))

(setq package-enable-at-startup nil)
(package-initialize)

saptarshiguha

---
title: Detecting Changes in Histograms
description: |
    Detecting Useful Changes in Histograms with small to many bins
author:
  - name: Saptarshi Guha
    affiliation: Product Metrics
date: "`r Sys.Date()`"
output: 
  radix::radix_article:

saptarshiguha

---
title: Have we missed profiles on ESR and Linux?
author: Saptarshi Guha <[email protected]>
date: "`r format(Sys.time(), '%H:%M %B %d, %Y',tz='America/Los_Angeles',usetz=TRUE)`"
output:
 html_document:
    mathjax: default
    self_contained: false
    theme: readable
    highlight: haddock

saptarshiguha

```{r}
releases <- local({
    x <- fromJSON(file='https://product-details.mozilla.org/1.0/firefox_history_major_releases.json')
    f <- rbindlist(Map(function(a,b){
        data.table(version=a, from=b)
    },names(x), x))
    f$from<- as.Date(f$from)
    f$to <- c(tail(f$from,-1)-1,as.Date('2025-01-01'))
    f
    })

saptarshiguha

## For Session Hours Per Day Per Profile

### For profiles on 56

with a as (
select client_id, submission_date_s3, sum(subsession_length)/3600 as hours
from main_summary
where app_name='Firefox'
and normalized_channel='release'
and substring(app_version,1,2)='56'

saptarshiguha

## For Session Hours Per Day Per Profile

### For profiles on 56

with a as (
select client_id, submission_date_s3, sum(subsession_length)/3600 as hours
from main_summary
where app_name='Firefox'
and substring(app_version,1,2)='56'
and submission_date_s3 >= '20170925'