bt2
AnasAbdin
Today's Document
let's talk about Bridgerton tea, my ask is open

★
Game of Thrones Daily

Love Begins

Janaina Medeiros
No title available
Sweet Seals For You, Always

PR's Tumblrdome

❣ Chile in a Photography ❣

No title available

izzy's playlists!
almost home
I'd rather be in outer space 🛸

oozey mess

Product Placement
NASA

#extradirty
Alisa U Zemlji Chuda

seen from Türkiye
seen from Indonesia

seen from Malaysia
seen from T1
seen from United States

seen from Malaysia

seen from T1

seen from United States
seen from Malaysia

seen from France
seen from Switzerland

seen from T1
seen from T1

seen from Germany

seen from Malaysia

seen from France
seen from United States
seen from United Arab Emirates

seen from China
seen from United Kingdom
@dmclenoir
bt2
bt2
bt2
bt2
bt2
bt2
bt2
bt2
bt2
bt2
bt2
bt2
Why Microsoft Has So Many Consumer Fails
I’m a glutton for punishment.
Zune. Microsoft Band V1. Windows Phone 7. You name the Microsoft consumer product and I’ve been an early adopter.
I’ve invested in learning the Microsoft enterprise/server stack and when I see an opportunity to leverage what I know about Microsoft technology in the business world with my consumer tech life I jump in with both feet.
But it never works. I just bought a Lumia 950 XL dual sim Windows phone. I owned the Lumia 1520 and was well versed in all the downsides of owning a Windows phone (sparse app store, no “ecosystem”, etc.).
Even though the case for Windows Phone hasn’t gotten much better and my early experiences with Windows 10 Mobile Software beta builds were poor, I felt like I was mentally prepared to deal with the roller coaster emotional ride of Windows phone ownership.
It was worse than I thought. It started with the Thanksgiving night line to purchase a phone from my local store (the Microsoft retail experience is worthy of a separate rant in and of itself).
Since that day I’ve battled uninformed retail and online store reps, a faulty Lumia Offers app, multiple phone resets, incomplete data transfer, and finger pointing between AT&T and Microsoft regarding phone configuration issues.
I’m now left with a phone that is less functional (no 4G LTE, no Visual Voicemail) than the 2-year-old phone that I had before with no ETA for when it’ll get better. Then it dawned on me. This is exactly what it’s like in the server space with Microsoft.
How many times have I been an early adopter of some new Microsoft technology (System Center, W2K8 R2, Azure early days) only to find that the tech introduced instability in my current setup or made my setup less functional.
Microsoft usually communicated pretty poorly about the issues initially and I had to rely on Technet, Microsoft MVP’s and support boards to supply answers while I waited for Microsoft to belatedly acknowledge a problem and even more belatedly provide a fix.
I’ve gotten used to it as an IT professional. And I thought I’d gotten used to dealing with it as an early adopter of Microsoft consumer technology. But I’m over it.
The issue is that the people in consumer business leadership positions at Microsoft are the same super smart people that presided over that server stack world and that they just don’t get the difference. This is why there are so many consumer fails.
Microsoft releases enterprise products and counts on communities like the MVP’s and the support boards to pick up the slack. As an IT professional I know this and set my expectations accordingly.
But we know that consumer experiences can and should be better. I’m not an Apple lover and lord knows I’ve had my share of poor Apple experiences, but I’ve never felt like I was “on my own” after purchasing an Apple product.
Microsoft, on the other hand, has totally abdicated consumer product support and problem resolution to the Windows blogger community.
I live on WindowsCentral and Thurrott.com constantly searching for solutions to my Surface Book and 950 XL problems, while Microsoft communicates poorly (Surface Book) or not all (950 XL Visual Voicemail).
And before people tell me that it’s not all Microsoft’s fault, I have to believe that Microsoft knows that as a consumer when I buy something with a Microsoft logo, even though there are tons of drivers (Intel HD Display for example) and services (AT&T Wireless service) that Microsoft doesn’t control, I hold Microsoft accountable. And you should too.
In so many areas of Microsoft’s consumer product process they’ve appropriated their enterprise approach to driving consumer engagement. They will continue to fail in the consumer space as long as they keep doing that.
I’m unlikely to leave the Microsoft server stack family. I’ve invested a ton of time learning and understanding how it works and I accept it and all its foibles.
But I don’t have to accept it anymore on the consumer side and the 950 XL experience might just be the proverbial straw that broke the camel’s back.
I also collapsed the answers of the “It’d be great to get pregnant right now” into two categories, Bad Choice = 1 and Good Choice = 0. I wanted to see if Black respondents were more likely to feel like getting pregnant now wouldn’t be so bad (keeping in mind that the average age of respondents is 17 (Assumes the survey was given in 1996)).
The takeaway is that Black respondents were more likely than other ethnic groups to be OK with getting pregnant now. Disappointing if correct.
Coursera Assignment 4 - Creating Graphs for Data
I’ve chosen the ADDHEALTH dataset. For the purposes of this assignment I chose two categorical variables: Ethnicity (HIGI9) and a question that asked about attitudes towards getting pregnant (HIRP2).
I renamed HIGI9 to ETH and renamed the number responses to the surveyors observed ethnicities (Black, White, American Indian, Asian, and eliminated all the other categories). I renamed HIRP2 to BAD (A respondent could answer the question It wouldn't be that bad if you got pregnant right now with a strongly agree to strongly disagree range. The average respondent age was 17 so I classified responses that said strongly agree/agree to Bad responses and anything else to good responses)
My code is below:
import pandas import numpy import seaborn import matplotlib.pyplot as plt
data = pandas.read_csv("addhealth_pds.csv", low_memory = False) data2 = data.copy() data2["H1GI1Y"] = data2["H1GI1Y"].convert_objects(convert_numeric = True) recode2 = {1: "White", 2:"Black", 3: "Indian", 4: "Asian", 5:"Other", 6:"Other", 7: "Other", 8:"Other", 9:"Other"} data2["ETH"] = data2["H1GI9"] data2["ETH"] = data2["ETH"].astype("category") data2["ETH"] = data2["ETH"].map(recode2) data2["ETH"] = data2["ETH"].replace(5, numpy.nan) data2["ETH"] = data2["ETH"].replace(6, numpy.nan) data2["ETH"] = data2["ETH"].replace(7, numpy.nan) data2["ETH"] = data2["ETH"].replace(8, numpy.nan) data2["ETH"] = data2["ETH"].replace(9, numpy.nan) seaborn.countplot(x= "ETH", data=data2) plt.xlabel("Observed Ethnicity of Respondents - Categorical") plt.ylabel("Count of Observed Ethnicities") plt.title("Distribution of Ethnicities")
data2["BAD"] = data2["H1RP2"] #data2["BAD"] = data2["BAD"].convert_objects(convert_numeric = True) data2["BAD"] = data2["BAD"].replace(6, numpy.nan) data2["BAD"] = data2["BAD"].replace(7, numpy.nan) recode1 = {1: "Pregnant Great!", 2:"Pregnant OK", 3: "50-50", 4: "Pregnant Bad", 5:"Pregnant Horrible"} data2["BAD"]=data2["BAD"].map(recode1) seaborn.countplot(x= "BAD", data=data2) plt.xlabel("Fear of Pregnancy") plt.title("Distribution of Bad Sexual Behavior Responses")
import pandas import numpy import seaborn import matplotlib.pyplot as plt
data = pandas.read_csv("addhealth_pds.csv", low_memory = False) data2 = data.copy() data2["H1GI1Y"] = data2["H1GI1Y"].convert_objects(convert_numeric = True) recode2 = {1: "Other", 2:"Black", 3: "Other", 4: "Other", 5:"Other", 6:"Other", 7: "Other", 8:"Other", 9:"Other"} data2["ETH"] = data2["H1GI9"] data2["ETH"] = data2["ETH"].astype("category") data2["ETH"] = data2["ETH"].map(recode2) data2["ETH"] = data2["ETH"].replace(5, numpy.nan) data2["ETH"] = data2["ETH"].replace(6, numpy.nan) data2["ETH"] = data2["ETH"].replace(7, numpy.nan) data2["ETH"] = data2["ETH"].replace(8, numpy.nan) data2["ETH"] = data2["ETH"].replace(9, numpy.nan)
data2["BAD"] = data2["H1RP2"].dropna().astype("category") data2["BAD"] = data2["BAD"].replace(6, numpy.nan) data2["BAD"] = data2["BAD"].replace(7, numpy.nan) recode1 = {1: "1", 2:"1",3: "0", 4: "0", 5:"0"} data2["BAD"]=data2["BAD"].map(recode1) data2["BAD"] = data2["BAD"].convert_objects(convert_numeric=True) seaborn.factorplot(x="ETH", y="BAD", data = data2, kind = "bar", ci=0) plt.xlabel("Ethnicity") plt.ylabel("Bad Sexual Attitudes")
Coursera Assignment 3
Latest assignment regarding managed variables. I used the ADDHEALTH dataset. I created managed variables for Gender, Ethnicity, and Birth Age of black female respondents. My code is first, then the output is next:
# -*- coding: utf-8 -*- """ Spyder Editor
This is a temporary script file. """ import pandas import numpy
data = pandas.read_csv('addhealth_pds.csv' , low_memory=False) data2 = data.copy() #data3=data2.copy() data2["BIO_SEX"] = data2["BIO_SEX"].convert_objects(convert_numeric = True) data2["H1GI6B"] = data2["H1GI6B"].convert_objects(convert_numeric = True) # Eliminate mis-entered Gender data data2["BIO_SEX"]=data["BIO_SEX"].replace(6, numpy.nan) recode1 = {1: "Male", 2: "Female"} data2["BIO_SEX"]=data2["BIO_SEX"].map(recode1) # Recode 1 and 2 to Male and Female respectively New_Sex = data2["BIO_SEX"].value_counts(sort=False, normalize=True) #print out new frequency distribution for Gender print(New_Sex) #Create recoding of race variable recode2 ={1: "African American", 0:"Other", 8:"Other"} data2["ETH"] = data2["H1GI6B"].map(recode2) #Create new variable name ethnicity = data2["ETH"].value_counts(sort=False, normalize =True) print(ethnicity) #Create new dataset that consists only Black Female survey respondents blkfem = data2[(data2["BIO_SEX"] == "Female") & (data2["H1GI6B"] == 1)] #Create a frequency distribution of the birth years of Black female respondents year =blkfem["H1GI1Y"].value_counts(sort=False, normalize=True) print(blkfem) print(year)
runfile('C:/Users/Don/Desktop/Coursera/Research Question - Relationship between Presence of Biological Father and Predisposition for Risky Sexual Behavior in Women/Assign3_DM.py', wdir='C:/Users/Don/Desktop/Coursera/Research Question - Relationship between Presence of Biological Father and Predisposition for Risky Sexual Behavior in Women')
(For Gender, changed 1 and 2 to Male and Female) Female 0.515990 Male 0.483856 dtype: float64
(For ethnicity, I combined the other entries for those who did not respond as African American into other) African American 0.248924 Other 0.750461 dtype: float64
(Created a new dataset consisting only of African American Female respondents and then calculated a frequency distribution of their ages)
74 0.004728 75 0.005910 76 0.053191 77 0.171395 78 0.171395 79 0.174941 80 0.182033 81 0.144208 82 0.089835 83 0.002364 dtype: float64
Coursera Assignment 2 : Python and Data Analysis
The objective for this lesson was to begin the process of setting up the analytics environment and learning enough Python and the fundamentals of IPython, Pandas, and NumPy to do some basic analytics. The results of the assignment are below:
My code is below and the output follows :
import pandas import numpy
data = pandas.read_csv('addhealth_pds.csv' , low_memory=False) #Prints the size of the data frame print(len(data)) print(len(data.columns))
#Converts all variables to numeric data["HIEDBLK"] = data["HIEDBLK"].convert_objects(convert_numeric = True) data["BIO_SEX"] = data["BIO_SEX"].convert_objects(convert_numeric = True) data["H1GI1Y"] = data["H1GI1Y"].convert_objects(convert_numeric = True)
#Prints counts and frequency tables for three variables print("First variable is Gender of respondent. A value of 1 implies a male respondent and value of 2 implies a female respondent. Any other value is interpreted as an error." ) sex_counts = data["BIO_SEX"].value_counts(sort=True) print (sex_counts)
print("Gender Frequency Table: A value of 1 implies a male respondent and value of 2 implies a female respondent. Any other value is interpreted as an error." ) sex_percent = data["BIO_SEX"].value_counts(sort=True, normalize=True) print (sex_percent)
print("Second variable is the Birth Year of the respondent. Variable H1GI1Y gives the 2 digit birth year of the respondent." ) age = data["H1GI1Y"].value_counts(sort=True, normalize=True) print (age)
print("Birth Year Frequency Table: Variable H1GI1Y gives the 2 digit birth year of the respondent." ) age_percent = data["H1GI1Y"].value_counts(sort=True, normalize=True) print (age_percent)
print("Final variable is the weighting percentage for Black Respondents. Only 520 of the respondents were black. To aid in the analytics, the study authors assigned a weighting value to each African American respondent.") BLK = data["HIEDBLK"].value_counts(sort = True, dropna=False) print (BLK)
print("Frequency table for weighting of Black respondents. All blank entries in the column were assigned NaN value") BLK_percent = data["HIEDBLK"].value_counts(sort=True, normalize=True, dropna=False) print (BLK_percent)
My output is
runfile('C:/Users/Don/Desktop/Coursera/Research Question - Relationship between Presence of Biological Father and Predisposition for Risky Sexual Behavior in Women/FirstTry.py', wdir='C:/Users/Don/Desktop/Coursera/Research Question - Relationship between Presence of Biological Father and Predisposition for Risky Sexual Behavior in Women')
6504 2829
First variable is Gender of respondent. A value of 1 implies a male respondent and value of 2 implies a female respondent. Any other value is interpreted as an error. 2 3356 1 3147 6 1 dtype: int64 Gender Frequency Table: A value of 1 implies a male respondent and value of 2 implies a female respondent. Any other value is interpreted as an error. 2 0.515990 1 0.483856 6 0.000154 dtype: float64 Second variable is the Birth Year of the respondent. Variable H1GI1Y gives the 2 digit birth year of the respondent. 78 0.179736 79 0.178659 77 0.176661 80 0.166205 81 0.138838 82 0.089022 76 0.059809 75 0.006611 74 0.002768 83 0.001230 96 0.000461 dtype: float64 Birth Year Frequency Table: Variable H1GI1Y gives the 2 digit birth year of the respondent. 78 0.179736 79 0.178659 77 0.176661 80 0.166205 81 0.138838 82 0.089022 76 0.059809 75 0.006611 74 0.002768 83 0.001230 96 0.000461 dtype: float64 Final variable is the weighting percentage for Black Respondents. Only 520 of the respondents were black. To aid in the analytics, the study authors assigned a weighting value to each African American respondent. NaN 5984 786.9350 64 1177.0300 32 1298.3324 24 941.0330 21 421.6026 21 783.6036 19 1667.5516 18 1643.7126 17 1773.9254 14 5709.3996 13 1614.6118 13 818.2474 13 1525.9212 13 1634.8698 9 1770.2374 9 1216.2932 9 1435.7202 8 1767.5004 8 638.9856 8 2443.4316 7 382.0028 7 1659.8194 7 963.5390 7 3969.8476 7 848.0896 7 1062.4796 6 942.3662 6 657.5244 6 604.1448 6 ... 6805.1906 2 3778.1564 2 1296.2396 2 1046.0984 2 1628.8596 2 1620.5734 2 838.2192 2 1530.4854 2 3496.7704 2 4171.9994 2 1185.5360 1 1060.7034 1 12340.6578 1 1167.3700 1 3135.9118 1 698.1540 1 1285.9974 1 1354.8448 1 679.7182 1 1446.2324 1 1456.4448 1 468.0972 1 493.4896 1 3893.3620 1 3643.4812 1 1630.4412 1 2785.6098 1 1745.5902 1 1697.7424 1 1612.0012 1 dtype: int64 Frequency table for weighting of Black respondents. All blank entries in the column were assigned NaN value NaN 0.920049 786.9350 0.009840 1177.0300 0.004920 1298.3324 0.003690 941.0330 0.003229 421.6026 0.003229 783.6036 0.002921 1667.5516 0.002768 1643.7126 0.002614 1773.9254 0.002153 5709.3996 0.001999 1614.6118 0.001999 818.2474 0.001999 1525.9212 0.001999 1634.8698 0.001384 1770.2374 0.001384 1216.2932 0.001384 1435.7202 0.001230 1767.5004 0.001230 638.9856 0.001230 2443.4316 0.001076 382.0028 0.001076 1659.8194 0.001076 963.5390 0.001076 3969.8476 0.001076 848.0896 0.001076 1062.4796 0.000923 942.3662 0.000923 657.5244 0.000923 604.1448 0.000923 ... 6805.1906 0.000308 3778.1564 0.000308 1296.2396 0.000308 1046.0984 0.000308 1628.8596 0.000308 1620.5734 0.000308 838.2192 0.000308 1530.4854 0.000308 3496.7704 0.000308 4171.9994 0.000308 1185.5360 0.000154 1060.7034 0.000154 12340.6578 0.000154 1167.3700 0.000154 3135.9118 0.000154 698.1540 0.000154 1285.9974 0.000154 1354.8448 0.000154 679.7182 0.000154 1446.2324 0.000154 1456.4448 0.000154 468.0972 0.000154 493.4896 0.000154 3893.3620 0.000154 3643.4812 0.000154 1630.4412 0.000154 2785.6098 0.000154 1745.5902 0.000154 1697.7424 0.000154 1612.0012 0.000154 dtype: float64