markdown stringlengths 0 37k | code stringlengths 1 33.3k | path stringlengths 8 215 | repo_name stringlengths 6 77 | license stringclasses 15
values |
|---|---|---|---|---|
Determine the number of aneurysms for which a variable is lower for CTA than for 3DRA. | numberofcases = np.empty(len(df_input.columns))
for i, variable in enumerate(df_input.columns):
numberofcases[i] = sum(df_3dra.loc[j, variable] > df_cta.loc[j, variable]
for j in df_input.index.levels[0])
s_numberofcases = pd.Series(numberofcases, index=df_input.columns) | data_analysis.ipynb | ajgeers/3dracta | bsd-2-clause |
Compose a dataframe with the obtained statistical results, corresponding to the 'online table' of the journal paper. | d = {'M': s_numberofcases,
'P': s_pvalue,
'Mean (%)': s_mean,
'SE (%)': s_standarderror}
df_output = pd.DataFrame(d, columns=['M', 'P', 'Mean (%)', 'SE (%)'])
df_output | data_analysis.ipynb | ajgeers/3dracta | bsd-2-clause |
Boxplot
Make boxplots showing the distributions of the relative differences over all aneurysms. | # extract arrays to plot from dataframe
array_yticklabels = ['$\mathregular{' + variable.replace('%', '\%') + '}$'
for variable in df_reldiff.columns]
array_reldiff = df_reldiff.values
# create plot
fig, ax = plt.subplots()
bp = ax.boxplot(array_reldiff, sym='+', vert=0, patch_artist=True)
# set ... | data_analysis.ipynb | ajgeers/3dracta | bsd-2-clause |
So basically here we are just flattening the images dimensions | # normalize
x_train /= 255
x_test /= 255
# flatten images
input_dim = 784
epochs = 20
# create training+test positive and negative pairs
digit_indices = [np.where(y_train == i)[0] for i in range(10)]
tr_pairs, tr_y = create_pairs(x_train, digit_indices)
digit_indices = [np.where(y_test == i)[0] for i in range(10)]
te... | Siamese Network - MNSIT - Breakdown.ipynb | gth158a/learning | apache-2.0 |
so this dimensions are basically:
+ 108400 is the number of items in the list
+ 2 each entry in the list has 2 parts (which are the pairs)
+ 784 is the dimension of the flatten image
scratch that. There is no list anymore since the function calls for np.array() before returning | tr_pairs[:,0].shape
tr_pairs[:,1].shape
tr_y.shape
len(digit_indices)
np.where(y_test == 2) | Siamese Network - MNSIT - Breakdown.ipynb | gth158a/learning | apache-2.0 |
np.where returns a list, so you need the [0] to get a array only | np.where(y_test == 2)[0] | Siamese Network - MNSIT - Breakdown.ipynb | gth158a/learning | apache-2.0 |
so the code above creates an array for each of the numbers, each array has the indices for each of the numbers
create_pairs function
python
def create_pairs(x, digit_indices):
'''Positive and negative pair creation.
Alternates between positive and negative pairs.
'''
pairs = []
labels = []
n = ... | digit_indices[0][2], digit_indices[0][2+1]
z1, z2 = digit_indices[0][2], digit_indices[0][2+1] | Siamese Network - MNSIT - Breakdown.ipynb | gth158a/learning | apache-2.0 |
so we are getting the indices of two intances of the same number | x_test[z1][:10]
x_test[z2][:10] | Siamese Network - MNSIT - Breakdown.ipynb | gth158a/learning | apache-2.0 |
then we are concatenating them | pairs += [[x_test[z1], x_test[z2]]]
len(pairs) | Siamese Network - MNSIT - Breakdown.ipynb | gth158a/learning | apache-2.0 |
so we keep building the pairs | inc = random.randrange(1, 10)
inc
d = 4
dn = (d + inc) % 10
dn | Siamese Network - MNSIT - Breakdown.ipynb | gth158a/learning | apache-2.0 |
here we are building a pair with a true and false instance thats why the inc on dn | # this adds the negative pair
# z1, z2 = digit_indices[d][i], digit_indices[dn][i]
# pairs += [[x[z1], x[z2]]] | Siamese Network - MNSIT - Breakdown.ipynb | gth158a/learning | apache-2.0 |
the label is 1 for the first example and zero for the fake that we just made | # labels += [1, 0]
pairs = []
labels = []
n = min([len(digit_indices[d]) for d in range(10)]) - 1
# how many indices are the per number
# get the min of all the indices
n
for d in range(10):
print(len(digit_indices[d]))
# then it substract 1 - I assume its because the range is non-inclusive
n = 10
ran = ... | Siamese Network - MNSIT - Breakdown.ipynb | gth158a/learning | apache-2.0 |
Variance
Once two statistician of height 4 feet and 5 feet have to cross a river of AVERAGE depth 3 feet. Meanwhile, a third person comes and said, "what are you waiting for? You can easily cross the river"
It's the average distance of the data values from the mean
<img style="float: left;" src="img/variance.png" hei... | shoes_before.head()
import matplotlib.pyplot as plt
%matplotlib inline
#Find the standard deviation of the sales
print("Before Campaign:")
print(shoes_before.std())
print()
print("During Campaign:")
print(shoes_during.std())
print()
print("After Campaign:")
print(shoes_after.std()) | Module_2f_ABTesting.ipynb | amitkaps/hackermath | mit |
Co-variance
covariance as a measure of the (average) co-variation between two variables, say x and y. Covariance describes both how far the variables are spread out, and the nature of their relationship, Covariance is a measure of how much two variables change together. Compare this to Variance, which is just the range... | #Find the covariance of the sales
#Use the cov function
print("Before Campaign:")
print(shoes_before.cov())
print()
print("During Campaign:")
print(shoes_during.cov())
print()
print("After Campaign:")
print(shoes_after.cov()) | Module_2f_ABTesting.ipynb | amitkaps/hackermath | mit |
Correlation
Extent to which two or more variables fluctuate together. A positive correlation indicates the extent to which those variables increase or decrease in parallel; a negative correlation indicates the extent to which one variable increases as the other decreases.
<img style="float: left;" src="img/correlation.... | import seaborn as sns
sns.jointplot(x= "UA101", y ="UA102", data=shoes_before)
sns.jointplot(x= "UA101", y ="UA102", data=shoes_during)
#Find correlation between sales
print("Before Campaign:")
print(shoes_before.corr())
print()
print("During Campaign:")
print(shoes_during.corr())
print()
print("After Campaign:")
pr... | Module_2f_ABTesting.ipynb | amitkaps/hackermath | mit |
Correlation != Causation
correlation between two variables does not necessarily imply that one causes the other.
<img style="float: left;" src="img/correlation_not_causation.gif" height="570" width="570"> | #Let's do some analysis on UA101 now
#Find difference between mean sales before and after the campaign
np.mean(shoes_after.UA101) - np.mean(shoes_before.UA101) | Module_2f_ABTesting.ipynb | amitkaps/hackermath | mit |
On average, the sales after campaign is more than the sales before campaign. But is the difference legit? Could it be due to chance?
Classical Method : t-test
Hacker's Method : provided in notebook 2c
Effect Size
Because you can't argue with all the fools in the world. It's easier to let them have their way, then tric... | #Find %increase in mean sales
(np.mean(shoes_after.UA101) - np.mean(shoes_before.UA101))/np.mean(shoes_after.UA101) * 100 | Module_2f_ABTesting.ipynb | amitkaps/hackermath | mit |
Would business feel comfortable spending millions of dollars if the increase is going to be 25%.
Does it work for the company?
Maybe yes - if margins are good and this increase is considered good. But if the returns from the campaign does not let the company break even, it makes no sense to take that path.
Someone t... | from scipy import stats
#Standard error for mean sales before campaign
stats.sem(shoes_before.UA101) | Module_2f_ABTesting.ipynb | amitkaps/hackermath | mit |
Hypothesis Testing
(We are covering, what is referred to as, frequentist method of Hypothesis testing)
We would like to know if the effects we see in the sample(observed data) are likely to occur in the population.
The way classical hypothesis testing works is by conducting a statistical test to answer the following q... | import numpy as np
import pandas as pd
from scipy import stats
import matplotlib as mpl
%matplotlib inline
import seaborn as sns
sns.set(color_codes=True)
#Mean of sales before campaign
shoes_before.UA101.mean()
#Confidence interval on the mean of sales before campaign
stats.norm.interval(0.95, loc=shoes_before.UA10... | Module_2f_ABTesting.ipynb | amitkaps/hackermath | mit |
Null Hypothesis: Mean sales aren't significantly different
Perform t-test and determine the p-value. | stats.ttest_ind(shoes_before.UA101,
shoes_after.UA101, equal_var=True) | Module_2f_ABTesting.ipynb | amitkaps/hackermath | mit |
p-value is the probability that the effective size was by chance. And here, p-value is almost 0.
Conclusion: The sales difference is significant.
Assumption of t-test
One assumption is that the data used came from a normal distribution.
<br>
There's a Shapiro-Wilk test to test for normality. If p-value is less than 0... | stats.shapiro(shoes_before.UA101)
?stats.shapiro | Module_2f_ABTesting.ipynb | amitkaps/hackermath | mit |
1.1
b. Find the average of a list of numbers using a for loop | runningTotal = 0
listOfNumbers = [4,7,9,1,8,6]
for each in listOfNumbers:
runningTotal = runningTotal + each
# each time round the loop add the next item to the running total
average = runningTotal/len(listOfNumbers)
# the average is the runningTotal at the end / how many numbers
print(listOfNumbers)
pr... | bicf_nanocourses/courses/python_1/lectures/day_2/Day2-Lesson-3-Exercise1_key.ipynb | bcantarel/bcantarel.github.io | gpl-3.0 |
1.1
c. Write a program that prints string in reverse. Print character by character
Input: Python
Expected Output:
n
o
h
t
y
P
Hint: can use range, len functions<br>
P y t h o n
0 1 2 3 4 5
-6 -5 -4 -3 -2 -1 | word = "Python"
#print(len(word))
for char in range(len(word) - 1, -1, -1): # range(start=5, end=-1, de -1)
print(word[char]) | bicf_nanocourses/courses/python_1/lectures/day_2/Day2-Lesson-3-Exercise1_key.ipynb | bcantarel/bcantarel.github.io | gpl-3.0 |
1.2
Write a Python program to count the number of even and odd numbers from a series of numbers.
<br>Hint: review of for loop, if statements | numbers = (1, 2, 3, 4, 5, 6, 7, 8, 9) # Declaring the tuple
count_odd = 0
count_even = 0
#type your code here
for x in numbers:
if not x % 2:
count_even+=1
else:
count_odd+=1
print("Number of even numbers :",count_even)
print("Number of odd numbers :",count_odd) | bicf_nanocourses/courses/python_1/lectures/day_2/Day2-Lesson-3-Exercise1_key.ipynb | bcantarel/bcantarel.github.io | gpl-3.0 |
1.3
Check if given list of strings have ECORI site motif and print value that doesn't contain the motif until two strings with the motif are found
motif = "GAATTC" (5' for ECORI restriction site)
Output:
<br>AGTGAACCGTCAGATCCGCTAGCGCGAATTC doesn't contain the motif
GGAGACCGACACCCTCCTGCTATGGGTGCTGCTGCTC doesn't contain... | motif = "GAATTC"
count = 0
dna_strings = ['AGTGAACCGTCAGATCCGCTAGCGCGAATTC','GGAGACCGACACCCTCCTGCTATGGGTGCTGCTGCTC','TGGGTGCCCGGCAGCACCGGCGACGCACCGGTCGC',
'CACCATGGTGAGCAAGGGCGAGGAGAATAACATGGCC','ATCATCAAGGAGTTCATGCGCTTCAAGAATTC','CATGGAGGGCTCCGTGAACGGCCACGAGTTCGAGA'
,'TCGAGGGCGAGGGCGAGGGC... | bicf_nanocourses/courses/python_1/lectures/day_2/Day2-Lesson-3-Exercise1_key.ipynb | bcantarel/bcantarel.github.io | gpl-3.0 |
1.4
Write a Python program that prints all the numbers in range from 0 to 10 except 5 and 10.
<br> Hint: use continue | #type your code here
for value in range(10):
if (value == 5 or value==10):
continue
print(value,end=' ')
print("\n") | bicf_nanocourses/courses/python_1/lectures/day_2/Day2-Lesson-3-Exercise1_key.ipynb | bcantarel/bcantarel.github.io | gpl-3.0 |
1.5 (Multi-Part)
Next series of tasks about lists and list manipulations:
<br>a. Create a list of 5 of your favorite things. | my_favorites=['Music', 'Movies', 'Coding', 'Biology', 'Python'] | bicf_nanocourses/courses/python_1/lectures/day_2/Day2-Lesson-3-Exercise1_key.ipynb | bcantarel/bcantarel.github.io | gpl-3.0 |
b. Use the print() function to print your list. | print(my_favorites) | bicf_nanocourses/courses/python_1/lectures/day_2/Day2-Lesson-3-Exercise1_key.ipynb | bcantarel/bcantarel.github.io | gpl-3.0 |
c. Use the print() function to print out the middle element. | print(my_favorites[2]) | bicf_nanocourses/courses/python_1/lectures/day_2/Day2-Lesson-3-Exercise1_key.ipynb | bcantarel/bcantarel.github.io | gpl-3.0 |
d. Now replace the middle element with a different item, your favorite song, or song bird. | my_favorites[2]='European robin' | bicf_nanocourses/courses/python_1/lectures/day_2/Day2-Lesson-3-Exercise1_key.ipynb | bcantarel/bcantarel.github.io | gpl-3.0 |
e. Use the same print statement from b. to print your new list. Check out the differences. | print(my_favorites) | bicf_nanocourses/courses/python_1/lectures/day_2/Day2-Lesson-3-Exercise1_key.ipynb | bcantarel/bcantarel.github.io | gpl-3.0 |
f. Add a new element to the end. Read about append(). | my_favorites.append('Monkeys') | bicf_nanocourses/courses/python_1/lectures/day_2/Day2-Lesson-3-Exercise1_key.ipynb | bcantarel/bcantarel.github.io | gpl-3.0 |
g. Add a new element to the beginning. Read about insert(). | my_favorites.insert(0, 'Evolution') | bicf_nanocourses/courses/python_1/lectures/day_2/Day2-Lesson-3-Exercise1_key.ipynb | bcantarel/bcantarel.github.io | gpl-3.0 |
h. Add a new element somewhere other than the beginning or the end. | my_favorites.insert(3, 'Coffee') | bicf_nanocourses/courses/python_1/lectures/day_2/Day2-Lesson-3-Exercise1_key.ipynb | bcantarel/bcantarel.github.io | gpl-3.0 |
1.6
Write a script that splits a string into a list:
Save the string sapiens, erectus, neanderthalensis as a variable.
Print the string.
Split the string into individual words and print the result of the split. (Think about the ', '.)
Store the resulting list in a new variable.
Print the list.
Sort the list alphabetic... | #type your code
hominins='sapiens, erectus, neanderthalensis'
print(hominins)
hominin_individuals=hominins.split(',')
print('hominin_individuals')
hominin_individuals=sorted(hominin_individuals)
print("List: ", hominin_individuals)
hominin_individuals=sorted(hominin_individuals, key=len)
print(hominin_individuals) | bicf_nanocourses/courses/python_1/lectures/day_2/Day2-Lesson-3-Exercise1_key.ipynb | bcantarel/bcantarel.github.io | gpl-3.0 |
Extra Pratice: 1.7
Use list comprehension to generate a list of tuples. The tuples should contain sequences and lengths from the previous problem. Print out the length and the sequence (i.e., "4\tATGC\n"). | sequences=['ATGCCCGGCCCGGC','GCGTGCTAGCAATACGATAAACCGG', 'ATATATATCGAT','ATGGGCCC']
seq_lengths=[(seq, len(seq)) for seq in sequences]
print(seq_lengths) | bicf_nanocourses/courses/python_1/lectures/day_2/Day2-Lesson-3-Exercise1_key.ipynb | bcantarel/bcantarel.github.io | gpl-3.0 |
Question 2: Dictionaries and Set
2.1
Create a dictionary store DNA restriction enzyme names and their motifs from:
<br>https://www.neb.com/tools-and-resources/selection-charts/alphabetized-list-of-recognition-specificities
<br>eg:
EcoRI = GAATTC
AvaII = GGACC
BisI = GGACC | enzymes = { 'EcoRI':'GAATTC','AvaII':'GGACC', 'BisI':'GCATGCGC' , 'SacII': r'CCGCGG','BamHI': 'GGATCC'}
print(enzymes) | bicf_nanocourses/courses/python_1/lectures/day_2/Day2-Lesson-3-Exercise1_key.ipynb | bcantarel/bcantarel.github.io | gpl-3.0 |
a. look up the motif for a particular SacII enzyme | print(enzymes['SacII']) | bicf_nanocourses/courses/python_1/lectures/day_2/Day2-Lesson-3-Exercise1_key.ipynb | bcantarel/bcantarel.github.io | gpl-3.0 |
b. add below two enzymes and their motifs to dictionary
<br>
KasI: GGCGCC
AscI: GGCGCGCC
EciI: GGCGGA | enzymes['KasI'] = 'GGCGCC'
enzymes['AscI'] = 'GGCGCGCC'
print(enzymes) | bicf_nanocourses/courses/python_1/lectures/day_2/Day2-Lesson-3-Exercise1_key.ipynb | bcantarel/bcantarel.github.io | gpl-3.0 |
2.2
Suppose dna is a string variable that contains only 'A','C','G' or 'T' characters.
Write code to find and print the character and frequency (max_freq) of the most frequent character in string dna?
<br> | dna = 'AAATTCGTGACTGTAA'
dna_counts= {'T':dna.count('T'),'C':dna.count('C'),'G':dna.count('G'),'A':dna.count('A')}
print(dna_counts)
max_freq= sorted(dna_counts.values())[-1]
print(max_freq) | bicf_nanocourses/courses/python_1/lectures/day_2/Day2-Lesson-3-Exercise1_key.ipynb | bcantarel/bcantarel.github.io | gpl-3.0 |
2.3
If you create a set using a DNA sequence, what will you get back? Try it with this sequence:
GATGGGATTGGGGTTTTCCCCTCCCATGTGCTCAAGACTGGCGCTAAAAGTTTTGAGCTTCTCAAAAGTCTAGAGCCACCGTCCAGGGAGCAGGTAGCTGCTGGGCTCCGGGGACACTTTGCGTTCGGGCTGGGAGCGTGCTTTCCACGACGGTGACACGCTTCCCTGGATTGGCAGCCAGACTGCCTTCCGGGTCACTGCCATGGAGGAGCCGCAGTCAGAT... | DNA='GATGGGATTGGGGTTTTCCCCTCCCATGTGCTCAAGACTGGCGCTAAAAGTTTTGAGCTTCTCAAAAGTCTAGAGCCACCGTCCAGGGAGCAGGTAGCTGCTGGGCTCCGGGGACACTTTGCGTTCGGGCTGGGAGCGTGCTTTCCACGACGGTGACACGCTTCCCTGGATTGGCAGCCAGACTGCCTTCCGGGTCACTGCCATGGAGGAGCCGCAGTCAGATCCTAGCGTCGAGCCCCCTCTGAGTCAGGAAACATTTTCAGACCTATGGAAACTACTTCCTGAAAACAACGTTCTGTCCCCCTTGCCGTCCCA... | bicf_nanocourses/courses/python_1/lectures/day_2/Day2-Lesson-3-Exercise1_key.ipynb | bcantarel/bcantarel.github.io | gpl-3.0 |
2. Scraping
The amount of insight you can gain by scraping free information from the net is astonishing, especially given how easy it is to scrape. Whatever you want to learn about, there is free information out there (and probably a Python package to help).
We will work on the simplest scenario: pulling off files that... | url = "http://bist.eu/"
req = Request(url, headers={'User-Agent': 'Mozilla/5.0'})
content = urlopen(req).read() | Archiv_Session_Spring_2017/Tutorials/Advanced_Data_Science.ipynb | peterwittek/qml-rg | gpl-3.0 |
If we inspect what it is, it looks like HTML: | content[:50] | Archiv_Session_Spring_2017/Tutorials/Advanced_Data_Science.ipynb | peterwittek/qml-rg | gpl-3.0 |
The way we read it, the content is a byte array: | type(content) | Archiv_Session_Spring_2017/Tutorials/Advanced_Data_Science.ipynb | peterwittek/qml-rg | gpl-3.0 |
Convert it to UTF-8 to see what it contains in a nicer way: | print(content.decode("utf-8")[:135]) | Archiv_Session_Spring_2017/Tutorials/Advanced_Data_Science.ipynb | peterwittek/qml-rg | gpl-3.0 |
You can save it to a file so that you do not have to scrape it again. This is important when you scrape millions of files, and the scraping has to be restarted. Have mercy on the webserver, or you risk getting blacklisted. | with open("bist.html", 'wb') as file:
file.write(content)
file.close() | Archiv_Session_Spring_2017/Tutorials/Advanced_Data_Science.ipynb | peterwittek/qml-rg | gpl-3.0 |
You can technically open this in a browser, but it is only the HTML part of the site: the style files, images, and a whole lot of other things that make a page working are missing. Lazy people can launch a browser from Python to see what the page looks like: | from subprocess import run
run(["firefox", "-new-tab", "bist.html"]) | Archiv_Session_Spring_2017/Tutorials/Advanced_Data_Science.ipynb | peterwittek/qml-rg | gpl-3.0 |
Exercise 1. Copy an image location from the BIST website. Scrape it and display it. For the display, you can import Image from IPython.display, and then use Image(data=content). Technically you could use imshow from Matplotlib, but then you would have decompress the image first.
We put everything together in a function... | def get_file(url, filename=None):
if filename is None:
slash = url.rindex("/")
filename = url[slash+1:]
if os.path.isfile(filename):
with open(filename, 'rb') as file:
content = file.read()
else:
req = Request(url, headers={'User-Agent': 'Mozilla/5.0'})
co... | Archiv_Session_Spring_2017/Tutorials/Advanced_Data_Science.ipynb | peterwittek/qml-rg | gpl-3.0 |
3. Cleaning structured data
We will integrate several data sources, as this is the most common pattern you encounter in day to day data science. We will have two structured sources of data, and one semistructured:
A table of journal titles and matching abbreviations. This will help us standardize journal titles in th... | abbreviations = get_file("https://github.com/JabRef/abbrv.jabref.org/raw/master/journals/journal_abbreviations_webofscience.txt") | Archiv_Session_Spring_2017/Tutorials/Advanced_Data_Science.ipynb | peterwittek/qml-rg | gpl-3.0 |
Let's see what it contains: | print(abbreviations.decode("utf-8")[:1000]) | Archiv_Session_Spring_2017/Tutorials/Advanced_Data_Science.ipynb | peterwittek/qml-rg | gpl-3.0 |
This is fairly straightforward: the file starts with a bunch of comments, then two columns containing the full title and the abbreviation, separated by the "=" character. We can load it with Pandas without thinking much: | abbs = pd.read_csv("journal_abbreviations_webofscience.txt", sep='=',
comment='#', names=["Full Journal Title", "Abbreviation"]) | Archiv_Session_Spring_2017/Tutorials/Advanced_Data_Science.ipynb | peterwittek/qml-rg | gpl-3.0 |
In principle, it looks okay: | abbs.head() | Archiv_Session_Spring_2017/Tutorials/Advanced_Data_Science.ipynb | peterwittek/qml-rg | gpl-3.0 |
If we take a closer look at a particular element, we will notice that something is off: | abbs.iloc[0, 1] | Archiv_Session_Spring_2017/Tutorials/Advanced_Data_Science.ipynb | peterwittek/qml-rg | gpl-3.0 |
The string starts with an unnecessary white space. We also do not want to miss any possible match because of differences in the use of capitalization. So we strip starting and trailing white space and convert every entry to upper case: | abbs = abbs.applymap(lambda x: x.upper().strip()) | Archiv_Session_Spring_2017/Tutorials/Advanced_Data_Science.ipynb | peterwittek/qml-rg | gpl-3.0 |
This is done: it was an easy and clean data set.
Let us go for the next one. Their Journal Citation Records reduced scientific contributions to a single number, allowing incompetent people to make unfounded judgements on the quality of your research output. So when you encounter a question like this: "I want to know wh... | get_file("https://www.researchgate.net/file.PostFileLoader.html?id=558730995e9d9735688b4631&assetKey=AS%3A273803718922244%401442291301717",
"2014_SCI_IF.xlsx"); | Archiv_Session_Spring_2017/Tutorials/Advanced_Data_Science.ipynb | peterwittek/qml-rg | gpl-3.0 |
We can open it in LibreOffice or OpenOffice, using the lazy way again: | run(["soffice", "2014_SCI_IF.xlsx"]) | Archiv_Session_Spring_2017/Tutorials/Advanced_Data_Science.ipynb | peterwittek/qml-rg | gpl-3.0 |
The two first lines are junk, and there is an index column. Let us load it in Pandas: | ifs = pd.read_excel("2014_SCI_IF.xlsx", skiprows=2, index_col=0) | Archiv_Session_Spring_2017/Tutorials/Advanced_Data_Science.ipynb | peterwittek/qml-rg | gpl-3.0 |
If this fails, you do not have the package for reading Excel files. Install it with import pip; pip.main(["install", "xlrd"]).
Now this table definitely looks fishy: | ifs.head() | Archiv_Session_Spring_2017/Tutorials/Advanced_Data_Science.ipynb | peterwittek/qml-rg | gpl-3.0 |
What are those empty columns? They do not seem to be in the file. The answer lies at the end of the table: | ifs.tail() | Archiv_Session_Spring_2017/Tutorials/Advanced_Data_Science.ipynb | peterwittek/qml-rg | gpl-3.0 |
Some stupid multi-column copyright notice at the bottom distorts the entire collection. This is why spreadsheets are the most hated data format for structured data, as the uncontrolled blend of content and formatting lets ignorant people give you a headache.
The journals are listed according to their ranking, in decrea... | skip = len(ifs) - ifs["Journal Impact Factor"].last_valid_index() + 1
ifs = pd.read_excel("2014_SCI_IF.xlsx", skiprows=2, index_col=0,
skip_footer=skip, parse_cols="A,B,E") | Archiv_Session_Spring_2017/Tutorials/Advanced_Data_Science.ipynb | peterwittek/qml-rg | gpl-3.0 |
This looks much better: | ifs.head() | Archiv_Session_Spring_2017/Tutorials/Advanced_Data_Science.ipynb | peterwittek/qml-rg | gpl-3.0 |
But not perfect: | ifs.sample(n=10) | Archiv_Session_Spring_2017/Tutorials/Advanced_Data_Science.ipynb | peterwittek/qml-rg | gpl-3.0 |
Now you see why we converted everything to upper case before.
Exercise 2. Convert everything in the column "Full Journal Title" to upper case. You will encounter a charming inconsistency in Pandas.
4. Cleaning and filtering of semi-structured data
The horror starts when even more control is given to humans who create t... | def get_search_result(author):
filename = author + ".html"
url = "https://arxiv.org/find/all/1/au:+" + author + "/0/1/0/all/0/1?per_page=400"
page = get_file(url, filename)
return BeautifulSoup(page, "html.parser") | Archiv_Session_Spring_2017/Tutorials/Advanced_Data_Science.ipynb | peterwittek/qml-rg | gpl-3.0 |
It is a common pattern to exploit search URLs to get what you want. Many don't let you pull off too many results in a single page, and you have to step through the "Next" links over and over again to get all the results you want. This is also not difficult, but it has the extra complication that you have to extract the... | lewenstein = get_search_result("Lewenstein_M") | Archiv_Session_Spring_2017/Tutorials/Advanced_Data_Science.ipynb | peterwittek/qml-rg | gpl-3.0 |
We can view the source in a browser: | run(["firefox", "-new-tab", "view-source:file://" + os.getcwd() + "/Lewenstein_M.html"]) | Archiv_Session_Spring_2017/Tutorials/Advanced_Data_Science.ipynb | peterwittek/qml-rg | gpl-3.0 |
Skipping the boring header part, the source code of the search result looks like this:
```html
<h3>Showing results 1 through 389 (of 389 total) for
<a href="/find/all/1/au:+Lewenstein_M/0/1/0/all/0/1?skip=0&query_id=ff0631708b5d0dd5">au:Lewenstein_M</a></h3>
<dl>
<dt>1. <span class="list-identifier"><a href="/abs... | titles = []
for dd in lewenstein.find_all("dd"):
titles.append(dd.find("div", class_="list-title mathjax"))
len(titles) | Archiv_Session_Spring_2017/Tutorials/Advanced_Data_Science.ipynb | peterwittek/qml-rg | gpl-3.0 |
So far so good, although the titles do not really look like we expect them: | titles[0] | Archiv_Session_Spring_2017/Tutorials/Advanced_Data_Science.ipynb | peterwittek/qml-rg | gpl-3.0 |
We define a helper function to extract the title: | def extract_title(title):
start = title.index(" ")
return title[start+1:-1]
titles = []
for dd in lewenstein.find_all("dd"):
titles.append(extract_title(dd.find("div", class_ = "list-title mathjax").text))
titles[0] | Archiv_Session_Spring_2017/Tutorials/Advanced_Data_Science.ipynb | peterwittek/qml-rg | gpl-3.0 |
The next problem we face is that not all of these papers belong to Maciej Lewenstein: some impostors have the same abbreviated name M. Lewenstein. They are easy to detect if they uses the non-abbreviated name. Let us run through the page again, noting which subject the the impostors publish in. For this, let us introdu... | def extract_subject(long_subject):
start = long_subject.index("(")
return long_subject[start+1:-1]
def drop_punctuation(string):
result = string.replace(".", " ")
return " ".join(result.split())
true_lewenstein = ["Maciej Lewenstein", "M Lewenstein"]
impostors = set()
primary_subjects = set()
for dd ... | Archiv_Session_Spring_2017/Tutorials/Advanced_Data_Science.ipynb | peterwittek/qml-rg | gpl-3.0 |
So it is only one person, and we can be reasonably confident that Maciej Lewenstein is unlikely to publish in these subjects. The other good news is that all the short forms of the name belong to physics papers, and not computer science. Armed with this knowledge, we can filter out the correct manuscripts.
We need to ... | def extract_journal(journal):
start = journal.index(" ")
raw = journal[start+1:-1]
m = re.search("\d", raw)
return drop_punctuation(raw[:m.start()]).strip().upper()
def extract_title(title):
start = title.index(" ")
return title[start+1:-1]
def extract_id_and_year(arXiv):
start = arXiv.i... | Archiv_Session_Spring_2017/Tutorials/Advanced_Data_Science.ipynb | peterwittek/qml-rg | gpl-3.0 |
We would be almost done if journal names were all entered same way. Of course they were not. Let us try to standardize them: this is where we first combine two sources of data. We use the journal abbreviations table to default to the long title of the journal in our paper collection. | for i, paper in enumerate(papers):
journal = paper[-1]
long_name = abbs[abbs["Abbreviation"] == journal]
if len(long_name) > 0:
papers[i][-1] = long_name["Full Journal Title"].values[0] | Archiv_Session_Spring_2017/Tutorials/Advanced_Data_Science.ipynb | peterwittek/qml-rg | gpl-3.0 |
There will still be some rotten apples: | def find_rotten_apples(paper_list):
rotten_apples = []
for paper in paper_list:
match = ifs[ifs["Full Journal Title"] == paper[-1]]
if len(match) == 0:
rotten_apples.append(paper[-1])
return sorted(rotten_apples)
rotten_apples = find_rotten_apples(papers)
rotten_apples | Archiv_Session_Spring_2017/Tutorials/Advanced_Data_Science.ipynb | peterwittek/qml-rg | gpl-3.0 |
Now you start to feel the pain of being a data scientist. The sloppiness of manual data entering is unbounded. Your duty is to clean up this mess. A quick fix is to tinker with the drop_punctuation function to replace the retarded encoding of JRC. Then go through the creation of the papers array and the standardization... | def drop_punctuation(string):
result = string.replace(".", " ")
result = result.replace(",", " ")
result = result.replace("(", " ")
result = result.replace(": ", "-")
return " ".join(result.split()) | Archiv_Session_Spring_2017/Tutorials/Advanced_Data_Science.ipynb | peterwittek/qml-rg | gpl-3.0 |
Exercise 3. Cut the number of rotten apples in half by defining a replacement dictionary and doing another round of standardization. | len(rotten_apples) | Archiv_Session_Spring_2017/Tutorials/Advanced_Data_Science.ipynb | peterwittek/qml-rg | gpl-3.0 |
It is the same drill with our other contender, except that the short version of his name is uniquely his. On the other hand, that single accent in the surname introduces N+1 spelling variants, which we should standardize. | acin = get_search_result("Acin_A")
for dd, dt in zip(acin.find_all("dd"), acin.find_all("dt")):
id_, year = extract_id_and_year(dt.find("a", attrs={"title": "Abstract"}).text)
div = dd.find("div", class_="list-authors")
subject = extract_subject(dd.find("span", class_ = "primary-subject").text)
journal ... | Archiv_Session_Spring_2017/Tutorials/Advanced_Data_Science.ipynb | peterwittek/qml-rg | gpl-3.0 |
Finally, we do another combination of sources. We merge the data set with the table of Impact Factors. The merge will be done based on the full journal titles. | db = pd.merge(pd.DataFrame(papers, columns=["arXiv", "Title", "Authors", "Primary Subject", "Year", "Full Journal Title"]),
ifs, how="inner", on=["Full Journal Title"]) | Archiv_Session_Spring_2017/Tutorials/Advanced_Data_Science.ipynb | peterwittek/qml-rg | gpl-3.0 |
Since the two of them co-authored papers, there are duplicated. They are easy to drop, since they have the same arXiv ID: | db = db.drop_duplicates(subset="arXiv") | Archiv_Session_Spring_2017/Tutorials/Advanced_Data_Science.ipynb | peterwittek/qml-rg | gpl-3.0 |
Notice that we lost a few papers: | print(len(papers), len(db)) | Archiv_Session_Spring_2017/Tutorials/Advanced_Data_Science.ipynb | peterwittek/qml-rg | gpl-3.0 |
This can be one of two reasons: the papers were published in journals that are not in JCR (very unlikely), or we failed matching the full journal name (very likely). It is getting tedious, so we live it here for now.
5. Visual analysis
Since we only focus on two authors, we can add an additional column to help identify... | def identify_key_authors(authors):
if "Maciej Lewenstein" in authors and "Antonio Acín" in authors:
return "AAML"
elif "Maciej Lewenstein" in authors:
return "ML"
else:
return "AA"
db["Group"] = db["Authors"].apply(lambda x: identify_key_authors(x)) | Archiv_Session_Spring_2017/Tutorials/Advanced_Data_Science.ipynb | peterwittek/qml-rg | gpl-3.0 |
Let's start plotting distributions: | groups = ["AA", "ML", "AAML"]
fig, ax = plt.subplots(ncols=1)
for group in groups:
data = db[db["Group"] == group]["Journal Impact Factor"]
sns.distplot(data, kde=False, label=group)
ax.legend()
ax.set_yscale("log")
plt.show() | Archiv_Session_Spring_2017/Tutorials/Advanced_Data_Science.ipynb | peterwittek/qml-rg | gpl-3.0 |
The logarithmic scale makes the raw number of papers appear more balanced, which is fair given the difference in age between the two authors. A single Nature paper makes a great outlier: | db[db["Journal Impact Factor"] == db["Journal Impact Factor"].max()] | Archiv_Session_Spring_2017/Tutorials/Advanced_Data_Science.ipynb | peterwittek/qml-rg | gpl-3.0 |
Actually, Toni has another Nature, but that is not on arXiv yet. Not all of Maciej's paper are on arXiv either, especially not the old ones.
We can do the same plots with subjects: | subjects = db["Primary Subject"].drop_duplicates()
fig, ax = plt.subplots(ncols=1)
for subject in subjects:
data = db[db["Primary Subject"] == subject]["Journal Impact Factor"]
sns.distplot(data, kde=False, label=subject)
ax.legend()
ax.set_yscale("log")
plt.show() | Archiv_Session_Spring_2017/Tutorials/Advanced_Data_Science.ipynb | peterwittek/qml-rg | gpl-3.0 |
You are safe with quant-ph and quantum gases, but stay clear of atom physics. It is amusing to restrict the histogram to Professor Acín's subset: | fig, ax = plt.subplots(ncols=1)
for subject in subjects:
data = db[(db["Primary Subject"] == subject) & (db["Group"] != "ML")]
if len(data) > 1:
sns.distplot(data["Journal Impact Factor"], kde=False, label=subject)
ax.legend()
ax.set_yscale("log")
plt.show() | Archiv_Session_Spring_2017/Tutorials/Advanced_Data_Science.ipynb | peterwittek/qml-rg | gpl-3.0 |
His topics are somewhat predictable.
Let's add one more column to indicate the number of authors: | db["#Authors"] = db["Authors"].apply(lambda x: len(x))
sns.stripplot(x="#Authors", y="Journal Impact Factor", data=db) | Archiv_Session_Spring_2017/Tutorials/Advanced_Data_Science.ipynb | peterwittek/qml-rg | gpl-3.0 |
How about the length of title? Number of words in title? | db["Length of Title"] = db["Title"].apply(lambda x: len(x))
db["Number of Words in Title"] = db["Title"].apply(lambda x: len(x.split()))
fig, axes = plt.subplots(ncols=2, figsize=(12, 5))
sns.stripplot(x="Length of Title", y="Journal Impact Factor", data=db, ax=axes[0])
sns.stripplot(x="Number of Words in Title", y="Jo... | Archiv_Session_Spring_2017/Tutorials/Advanced_Data_Science.ipynb | peterwittek/qml-rg | gpl-3.0 |
Do our authors maintain IF over time? | fig, axes = plt.subplots(nrows=2, figsize=(10, 5))
data = db[db["Group"] != "ML"]
sns.stripplot(x="Year", y="Journal Impact Factor", data=data, ax=axes[0])
axes[0].set_title("AA")
data = db[db["Group"] != "AA"].sort_values(by="Year")
sns.stripplot(x="Year", y="Journal Impact Factor", data=data, ax=axes[1])
axes[1].set_... | Archiv_Session_Spring_2017/Tutorials/Advanced_Data_Science.ipynb | peterwittek/qml-rg | gpl-3.0 |
We start by a simple hello world. This code plots the numbers $[0, 10)$ and adds a title Hello zmpl. Notice you don't need to call zmpl.show to see the results. | zplt.clf()
zplt.plot(range(10))
zplt.title('Hello zmpl!') | examples/zmpl_demo.ipynb | willyd/zmpl | mit |
Now we can acheive some interactivity very simply without using the ion and pause functions. This code will generate a random image of 100 by 100 pixels and displays them using imshow. | # disable auto draw
zplt.auto_draw(False)
for i in range(10):
zplt.clf()
data = np.random.random((100, 100, 3))
zplt.hold(True)
zplt.imshow(data)
zplt.plot(range(25, 76), range(25, 76), 'k-', linewidth=2.0)
zplt.axis('square')
zplt.axis('tight')
zplt.hold(False)
zpl | examples/zmpl_demo.ipynb | willyd/zmpl | mit |
We can also use some module objects like figure and axes | # enable auto draw
zplt.auto_draw(True)
f1 = zplt.figure(1)
ax1 = f1.add_subplot(111)
ax1.plot(range(10), 'g-')
zplt.title('Figure 1')
f2 = zplt.figure(2)
ax2 = f2.add_subplot(111)
ax2.plot(range(20), 'r-')
zplt.title('Figure 2') | examples/zmpl_demo.ipynb | willyd/zmpl | mit |
The objects are special proxy types that are created on the fly by our simple RPC system. | print(type(f1))
print(dir(f1)[:20])
print(type(ax1))
print(dir(ax1)[:20]) | examples/zmpl_demo.ipynb | willyd/zmpl | mit |
1.2. BigQuery Storage & Spark SQL - Python
Create Dataproc Cluster with Jupyter
This notebook is designed to be run on Google Cloud Dataproc.
Follow the links below for instructions on how to create a Dataproc Cluster with the Juypter component installed.
Tutorial - Install and run a Jupyter notebook on a Dataproc clu... | !scala -version | notebooks/python/1.2. BigQuery Storage & Spark SQL - Python.ipynb | GoogleCloudDataproc/cloud-dataproc | apache-2.0 |
Create Spark Session
Include the correct version of the spark-bigquery-connector jar
If you are using scala version 2.11 use 'gs://spark-lib/bigquery/spark-bigquery-latest.jar'.
If you are using scala version 2.12 use 'gs://spark-lib/bigquery/spark-bigquery-latest_2.12.jar'. | from pyspark.sql import SparkSession
spark = SparkSession.builder \
.appName('1.2. BigQuery Storage & Spark SQL - Python')\
.config('spark.jars', 'gs://spark-lib/bigquery/spark-bigquery-latest.jar') \
.getOrCreate() | notebooks/python/1.2. BigQuery Storage & Spark SQL - Python.ipynb | GoogleCloudDataproc/cloud-dataproc | apache-2.0 |
Read BigQuery table into Spark Dataframe
Use filter() to query data from a partitioned table. | table = "bigquery-public-data.wikipedia.pageviews_2020"
df_wiki_pageviews = spark.read \
.format("bigquery") \
.option("table", table) \
.option("filter", "datehour >= '2020-03-01' AND datehour < '2020-03-02'") \
.load()
df_wiki_pageviews.printSchema() | notebooks/python/1.2. BigQuery Storage & Spark SQL - Python.ipynb | GoogleCloudDataproc/cloud-dataproc | apache-2.0 |
Create temp table
Create temp table to be used in Spark SQL queries | df_wiki_pageviews.createOrReplaceTempView("wiki_pageviews") | notebooks/python/1.2. BigQuery Storage & Spark SQL - Python.ipynb | GoogleCloudDataproc/cloud-dataproc | apache-2.0 |
Select required columns and apply a filter using WHERE | df_wiki_en = spark.sql("""
SELECT
title, wiki, views
FROM wiki_pageviews
WHERE views > 1000 AND wiki in ('en', 'en.m')
""").cache()
df_wiki_en | notebooks/python/1.2. BigQuery Storage & Spark SQL - Python.ipynb | GoogleCloudDataproc/cloud-dataproc | apache-2.0 |
Create a wiki en pageviews table | df_wiki_en.createOrReplaceTempView("wiki_en") | notebooks/python/1.2. BigQuery Storage & Spark SQL - Python.ipynb | GoogleCloudDataproc/cloud-dataproc | apache-2.0 |
Group by title and find top pages by page views | df_wiki_en_totals = spark.sql("""
SELECT
title,
SUM(views) as total_views
FROM wiki_en
GROUP BY title
ORDER BY total_views DESC
""")
df_wiki_en_totals | notebooks/python/1.2. BigQuery Storage & Spark SQL - Python.ipynb | GoogleCloudDataproc/cloud-dataproc | apache-2.0 |
By default, cross_val_score will use StratifiedKFold for classification, which ensures that the class proportions in the dataset are reflected in each fold. If you have a binary classification dataset with 90% of data point belonging to class 0, that would mean that in each fold, 90% of datapoints would belong to class... | cv = StratifiedKFold(n_splits=5)
for train, test in cv.split(X, y):
print(test) | Day_1_Scientific_Python/scikit-learn/13-Cross-Validation.ipynb | paris-saclay-cds/python-workshop | bsd-3-clause |
As you can see, there are a couple of samples from the beginning, then from the middle, and then from the end, in each of the folds.
This way, the class ratios are preserved. Let's visualize the split: | def plot_cv(cv, y):
masks = []
X = np.ones((len(y), 1))
for train, test in cv.split(X, y):
mask = np.zeros(len(y), dtype=bool)
mask[test] = 1
masks.append(mask)
plt.matshow(masks)
plot_cv(StratifiedKFold(n_splits=5), iris.target) | Day_1_Scientific_Python/scikit-learn/13-Cross-Validation.ipynb | paris-saclay-cds/python-workshop | bsd-3-clause |
For comparison, again the standard KFold, that ignores the labels: | plot_cv(KFold(n_splits=5), iris.target) | Day_1_Scientific_Python/scikit-learn/13-Cross-Validation.ipynb | paris-saclay-cds/python-workshop | bsd-3-clause |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.