markdown
stringlengths
0
1.02M
code
stringlengths
0
832k
output
stringlengths
0
1.02M
license
stringlengths
3
36
path
stringlengths
6
265
repo_name
stringlengths
6
127
Class RecommenderEvaluatorIn order to become easier to evaluate the metrics, I created a class that receives all the original ratings and predicted ratings for every recommender system and defined functions to extract all the metrics established in section 1 of the capstone report. Lets take a look at a summary of the...
class RecommenderEvaluator: def __init__(self, items, actual_ratings, content_based, user_user, item_item, matrix_fact, pers_bias): self.items = items self.actual_ratings = actual_ratings # static data containing the average score given by each user self.average_rating_...
_____no_output_____
MIT
notebooks/reco-tut-asr-99-10-metrics-calculation.ipynb
sparsh-ai/reco-tut-asr
Test methods:Just to have an idea of the output of each method, lets call all them with a test user. At the next section we will calculate these metrics for all users.
userId = '64' re = RecommenderEvaluator(items, actual_ratings, content_based, user_user, item_item, matrix_fact, pers_bias)
_____no_output_____
MIT
notebooks/reco-tut-asr-99-10-metrics-calculation.ipynb
sparsh-ai/reco-tut-asr
Test RMSE
re.rmse(userId)
_____no_output_____
MIT
notebooks/reco-tut-asr-99-10-metrics-calculation.ipynb
sparsh-ai/reco-tut-asr
Test nDCG
re.nDCG(userId)
_____no_output_____
MIT
notebooks/reco-tut-asr-99-10-metrics-calculation.ipynb
sparsh-ai/reco-tut-asr
Test Diversity - Price and Availability
re.price_diversity(userId) re.availability_diversity(userId)
_____no_output_____
MIT
notebooks/reco-tut-asr-99-10-metrics-calculation.ipynb
sparsh-ai/reco-tut-asr
Test Popularity
re.popularity(userId)
_____no_output_____
MIT
notebooks/reco-tut-asr-99-10-metrics-calculation.ipynb
sparsh-ai/reco-tut-asr
Test Precision@N
re.precision_at_n(userId)
_____no_output_____
MIT
notebooks/reco-tut-asr-99-10-metrics-calculation.ipynb
sparsh-ai/reco-tut-asr
Average metrics by all usersEspefically for user 907, the recommendations from the user user came with all nulls (original dataset). This specifically impacted the RMSE calculation, as one Nan damaged the entire average calculation. So specifically for RMSE we did a separate calculation section. All the other metrics ...
re = RecommenderEvaluator(items, actual_ratings, content_based, user_user, item_item, matrix_fact, pers_bias) i = 0 count = np.array([0,0,0,0,0]) for userId in actual_ratings.columns: if(userId == '907'): rmse_recommenders = re.rmse(userId).fillna(0) else: rmse_recommenders = re.rmse(userId) ...
--- Average nDCG --- content_based 0.136505 item_item 0.146798 matrix_fact 0.155888 pers_bias 0.125180 user_user 0.169080 Name: nDCG, dtype: float64 --- Average Price - Diversity Measure --- mean std content_based 19.286627 19.229536 user_user 21.961776...
MIT
notebooks/reco-tut-asr-99-10-metrics-calculation.ipynb
sparsh-ai/reco-tut-asr
Final AnalysisIn terms of **RMSE**, the user-user collaborative filtering showed to be the most effective, despite it not being significantly better.For nDCG rank score, again user user and now item item collaborative filtering were the best.In terms of price diversity, the item item algorith was the most diverse, pro...
obs_ratings_list = [] content_based_list = [] user_user_list = [] item_item_list = [] matrix_fact_list = [] pers_bias_list = [] re = RecommenderEvaluator(items, actual_ratings, content_based, user_user, item_item, matrix_fact, pers_bias) for userId in actual_ratings.columns: observed_ratings = re.get_observed...
_____no_output_____
MIT
notebooks/reco-tut-asr-99-10-metrics-calculation.ipynb
sparsh-ai/reco-tut-asr
In order to have an idea of the results, let's choose 3 users randomly to show the predictions using the new hybrid models
np.random.seed(42) sample_users = np.random.choice(actual_ratings.columns, 3).astype(str) print('sample_users: ' + str(sample_users))
sample_users: ['1528' '3524' '417']
MIT
notebooks/reco-tut-asr-99-10-metrics-calculation.ipynb
sparsh-ai/reco-tut-asr
Get recommenders' predictions for sample users in order to create input for ensemble models (hybridization I and II)
from collections import OrderedDict df_sample = pd.DataFrame() for user in sample_users: content_based_ = re.content_based[user] user_user_ = re.user_user[user] item_item_ = re.item_item[user] matrix_fact_ = re.matrix_fact[user] pers_bias_ = re.pers_bias[user] df_sample = df_sample.append(pd.Da...
_____no_output_____
MIT
notebooks/reco-tut-asr-99-10-metrics-calculation.ipynb
sparsh-ai/reco-tut-asr
Focus on Performance (RMSE) I - Linear Model
from sklearn.linear_model import LinearRegression from sklearn.model_selection import cross_val_score linear = LinearRegression() print('RMSE for linear ensemble of recommender systems:') np.mean(cross_val_score(linear, dataset.drop('rating', axis=1), dataset['rating'], cv=5))
RMSE for linear ensemble of recommender systems:
MIT
notebooks/reco-tut-asr-99-10-metrics-calculation.ipynb
sparsh-ai/reco-tut-asr
Predictions for sample users: Creating top 5 recommendations for sample users
pred_cols = ['content_based','user_user','item_item','matrix_fact','pers_bias'] predictions = linear.fit(dataset.drop('rating', axis=1), dataset['rating']).predict(df_sample[pred_cols]) recommendations = pd.DataFrame(OrderedDict({'user':df_sample['user'], 'item':df_sample['item'], 'predictions':predictions})) recommend...
_____no_output_____
MIT
notebooks/reco-tut-asr-99-10-metrics-calculation.ipynb
sparsh-ai/reco-tut-asr
Focus on Performance (RMSE) II - Emsemble
from sklearn.ensemble import RandomForestRegressor rf = RandomForestRegressor(random_state=42) print('RMSE for non linear ensemble of recommender systems:') np.mean(cross_val_score(rf, dataset.drop('rating', axis=1), dataset['rating'], cv=5))
RMSE for non linear ensemble of recommender systems:
MIT
notebooks/reco-tut-asr-99-10-metrics-calculation.ipynb
sparsh-ai/reco-tut-asr
Predictions for sample users:
predictions = rf.fit(dataset.drop('rating', axis=1), dataset['rating']).predict(df_sample[pred_cols]) recommendations = pd.DataFrame(OrderedDict({'user':df_sample['user'], 'item':df_sample['item'], 'predictions':predictions})) recommendations.groupby('user').apply(lambda df_user : df_user.loc[df_user['predictions'].sor...
_____no_output_____
MIT
notebooks/reco-tut-asr-99-10-metrics-calculation.ipynb
sparsh-ai/reco-tut-asr
Focus on Recommendations - Top 1 from each RecommenderWith the all top 1 recommender, we can evaluate its performance not just with RMSE, but all the list metrics we evaluated before. As a business constraint, we will also pay more attention to the *precision@5* metric, as a general information on how good is the reco...
count_nDCG = np.array([0]) count_diversity_price = np.ndarray([1,2]) count_diversity_availability = np.ndarray([1,2]) count_popularity = np.array([0]) count_precision = np.array([0]) for userId in actual_ratings.columns: top_n_1 = re.get_top_n(userId,1) user_items = {} user_items['top_1_all'] = [a[0] ...
--- Average nDCG --- top_1_all 0.159211 Name: nDCG, dtype: float64 --- Average Price - Diversity Measure --- mean std top_1_all 16.4625 14.741783 --- Average Availability - Diversity Measure --- mean std top_1_all 0.575683 0.161168 --- Average Popularity --- top_...
MIT
notebooks/reco-tut-asr-99-10-metrics-calculation.ipynb
sparsh-ai/reco-tut-asr
Predictions for sample users:
results = {} for user_sample in sample_users: results[user_sample] = [a[0] for a in list(re.get_top_n(user_sample, 1).values())] results
_____no_output_____
MIT
notebooks/reco-tut-asr-99-10-metrics-calculation.ipynb
sparsh-ai/reco-tut-asr
Focus on Recommendations - Switching algorithm Can we use a Content Based Recommender for items with less evaluations?We can see in the cumulative histogram that only around 20% of the rated items had 10 or more ratings. This signals us that maybe we can prioritize the use of a content based recommender or even a non...
import matplotlib.pyplot as plt item_nbr_ratings = actual_ratings.apply(lambda col: np.sum(~np.isnan(col)), axis=1) item_max_nbr_ratings = item_nbr_ratings.max() range_item_max_nbr_ratings = range(item_max_nbr_ratings+1) plt.figure(figsize=(15,3)) plt.subplot(121) nbr_ratings_items = [] for i in range_item_max_nbr_ra...
_____no_output_____
MIT
notebooks/reco-tut-asr-99-10-metrics-calculation.ipynb
sparsh-ai/reco-tut-asr
Geometric operations Overlay analysisIn this tutorial, the aim is to make an overlay analysis where we create a new layer based on geometries from a dataset that `intersect` with geometries of another layer. As our test case, we will select Polygon grid cells from `TravelTimes_to_5975375_RailwayStation_Helsinki.shp` t...
import geopandas as gpd import matplotlib.pyplot as plt import shapely.speedups %matplotlib inline # File paths border_fp = "data/Helsinki_borders.shp" grid_fp = "data/TravelTimes_to_5975375_RailwayStation.shp" # Read files grid = gpd.read_file(grid_fp) hel = gpd.read_file(border_fp)
_____no_output_____
MIT
geometric-operations.ipynb
AdrianKriger/Automating-GIS-Processess
- Visualize the layers:
# Plot the layers ax = grid.plot(facecolor='gray') hel.plot(ax=ax, facecolor='None', edgecolor='blue')
_____no_output_____
MIT
geometric-operations.ipynb
AdrianKriger/Automating-GIS-Processess
Here the grey area is the Travel Time Matrix grid (13231 grid squares) that covers the Helsinki region, and the blue area represents the municipality of Helsinki. Our goal is to conduct an overlay analysis and select the geometries from the grid polygon layer that intersect with the Helsinki municipality polygon.When c...
# Ensure that the CRS matches, if not raise an AssertionError assert hel.crs == grid.crs, "CRS differs between layers!"
_____no_output_____
MIT
geometric-operations.ipynb
AdrianKriger/Automating-GIS-Processess
Indeed, they do. Hence, the pre-requisite to conduct spatial operations between the layers is fullfilled (also the map we plotted indicated this).- Let's do an overlay analysis and create a new layer from polygons of the grid that `intersect` with our Helsinki layer. We can use a function called `overlay()` to conduct ...
intersection = gpd.overlay(grid, hel, how='intersection')
_____no_output_____
MIT
geometric-operations.ipynb
AdrianKriger/Automating-GIS-Processess
- Let's plot our data and see what we have:
intersection.plot(color="b")
_____no_output_____
MIT
geometric-operations.ipynb
AdrianKriger/Automating-GIS-Processess
As a result, we now have only those grid cells that intersect with the Helsinki borders. As we can see **the grid cells are clipped based on the boundary.**- Whatabout the data attributes? Let's see what we have:
print(intersection.head())
car_m_d car_m_t car_r_d car_r_t from_id pt_m_d pt_m_t pt_m_tt \ 0 29476 41 29483 46 5876274 29990 76 95 1 29456 41 29462 46 5876275 29866 74 95 2 36772 50 36778 56 5876278 33541 116 137 3 36898 49 ...
MIT
geometric-operations.ipynb
AdrianKriger/Automating-GIS-Processess
As we can see, due to the overlay analysis, the dataset contains the attributes from both input layers.- Let's save our result grid as a GeoJSON file that is commonly used file format nowadays for storing spatial data.
# Output filepath outfp = "data/TravelTimes_to_5975375_RailwayStation_Helsinki.geojson" # Use GeoJSON driver intersection.to_file(outfp, driver="GeoJSON")
_____no_output_____
MIT
geometric-operations.ipynb
AdrianKriger/Automating-GIS-Processess
There are many more examples for different types of overlay analysis in [Geopandas documentation](http://geopandas.org/set_operations.html) where you can go and learn more. Aggregating dataData aggregation refers to a process where we combine data into groups. When doing spatial data aggregation, we merge the geometri...
# Conduct the aggregation dissolved = intersection.dissolve(by="car_r_t") # What did we get print(dissolved.head())
geometry car_m_d car_m_t \ car_r_t -1 MULTIPOLYGON (((388000.000 6668750.000, 387750... -1 -1 0 POLYGON ((386000.000 6672000.000, 385750.000 6... 0 0 ...
MIT
geometric-operations.ipynb
AdrianKriger/Automating-GIS-Processess
- Let's compare the number of cells in the layers before and after the aggregation:
print('Rows in original intersection GeoDataFrame:', len(intersection)) print('Rows in dissolved layer:', len(dissolved))
Rows in original intersection GeoDataFrame: 3826 Rows in dissolved layer: 51
MIT
geometric-operations.ipynb
AdrianKriger/Automating-GIS-Processess
Indeed the number of rows in our data has decreased and the Polygons were merged together.What actually happened here? Let's take a closer look. - Let's see what columns we have now in our GeoDataFrame:
print(dissolved.columns)
Index(['geometry', 'car_m_d', 'car_m_t', 'car_r_d', 'from_id', 'pt_m_d', 'pt_m_t', 'pt_m_tt', 'pt_r_d', 'pt_r_t', 'pt_r_tt', 'to_id', 'walk_d', 'walk_t', 'GML_ID', 'NAMEFIN', 'NAMESWE', 'NATCODE'], dtype='object')
MIT
geometric-operations.ipynb
AdrianKriger/Automating-GIS-Processess
As we can see, the column that we used for conducting the aggregation (`car_r_t`) can not be found from the columns list anymore. What happened to it?- Let's take a look at the indices of our GeoDataFrame:
print(dissolved.index)
Int64Index([-1, 0, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 56], dtype='int64', name='car_r_t')
MIT
geometric-operations.ipynb
AdrianKriger/Automating-GIS-Processess
Aha! Well now we understand where our column went. It is now used as index in our `dissolved` GeoDataFrame. - Now, we can for example select only such geometries from the layer that are for example exactly 15 minutes away from the Helsinki Railway Station:
# Select only geometries that are within 15 minutes away dissolved.iloc[15] # See the data type print(type(dissolved.iloc[15])) # See the data print(dissolved.iloc[15].head())
geometry (POLYGON ((388250.0001354316 6668750.000042891... car_m_d 12035 car_m_t 18 car_r_d 11997 from_id 5903886 Name: 20, ...
MIT
geometric-operations.ipynb
AdrianKriger/Automating-GIS-Processess
As we can see, as a result, we have now a Pandas `Series` object containing basically one row from our original aggregated GeoDataFrame.Let's also visualize those 15 minute grid cells.- First, we need to convert the selected row back to a GeoDataFrame:
# Create a GeoDataFrame selection = gpd.GeoDataFrame([dissolved.iloc[15]], crs=dissolved.crs)
_____no_output_____
MIT
geometric-operations.ipynb
AdrianKriger/Automating-GIS-Processess
- Plot the selection on top of the entire grid:
# Plot all the grid cells, and the grid cells that are 15 minutes a way from the Railway Station ax = dissolved.plot(facecolor='gray') selection.plot(ax=ax, facecolor='red')
_____no_output_____
MIT
geometric-operations.ipynb
AdrianKriger/Automating-GIS-Processess
Simplifying geometries Sometimes it might be useful to be able to simplify geometries. This could be something to consider for example when you have very detailed spatial features that cover the whole world. If you make a map that covers the whole world, it is unnecessary to have really detailed geometries because it ...
import geopandas as gpd # File path fp = "data/Amazon_river.shp" data = gpd.read_file(fp) # Print crs print(data.crs) # Plot the river data.plot();
PROJCS["Mercator_2SP",GEOGCS["GCS_GRS 1980(IUGG, 1980)",DATUM["D_unknown",SPHEROID["GRS80",6378137,298.257222101]],PRIMEM["Unknown",0],UNIT["Degree",0.0174532925199433]],PROJECTION["Mercator_2SP"],PARAMETER["standard_parallel_1",-2],PARAMETER["central_meridian",-43],PARAMETER["false_easting",5000000],PARAMETER["false_n...
MIT
geometric-operations.ipynb
AdrianKriger/Automating-GIS-Processess
The LineString that is presented here is quite detailed, so let's see how we can generalize them a bit. As we can see from the coordinate reference system, the data is projected in a metric system using [Mercator projection based on SIRGAS datum](http://spatialreference.org/ref/sr-org/7868/). - Generalization can be do...
# Generalize geometry data2 = data.copy() data2['geom_gen'] = data2.simplify(tolerance=20000) # Set geometry to be our new simlified geometry data2 = data2.set_geometry('geom_gen') # Plot data2.plot() # plot them side-by-side %matplotlib inline import matplotlib.pyplot as plt #basic config fig, (ax1,ax2) = plt.subp...
_____no_output_____
MIT
geometric-operations.ipynb
AdrianKriger/Automating-GIS-Processess
GNN Implementation- Name: Abhishek Aditya BS- SRN: PES1UG19CS019- VI Semester 'A' Section- Date: 27-04-2022
import sys if 'google.colab' in sys.modules: %pip install -q stellargraph[demos]==1.2.1 import pandas as pd import os import stellargraph as sg from stellargraph.mapper import FullBatchNodeGenerator from stellargraph.layer import GCN from tensorflow.keras import layers, optimizers, losses, metrics, Model from sklear...
_____no_output_____
MIT
Topics of Deep Learning Lab/Lab-3/GNN.ipynb
Abhishek-Aditya-bs/Lab-Projects-and-Assignments
Within repo
import pandas as pd train_file = ["mesos", "usergrid", "appceleratorstudio", "appceleratorstudio", "titanium", "aptanastudio", "mule", "mulestudio"] test_file = ["usergrid", "mesos", "aptanastudio", "titanium", "appceleratorstudio", "titanium", "mulestudio", "mule"] mae = [1.07, 1.14, 2.75, 1.99, 2.85, 3.41, 3.14, 2.3...
_____no_output_____
MIT
abe0/ignore_process_csv.ipynb
awsm-research/gpt2sp
Cross repo
import pandas as pd train_file = ["clover", "talendesb", "talenddataquality", "mule", "talenddataquality", "mulestudio", "appceleratorstudio", "appceleratorstudio"] test_file = ["usergrid", "mesos", "aptanastudio", "titanium", "appceleratorstudio", "titanium", "mulestudio", "mule"] mae = [1.57, 2.08, 5.37, 6.36, 5.55,...
_____no_output_____
MIT
abe0/ignore_process_csv.ipynb
awsm-research/gpt2sp
Solution to puzzle number 5
import pandas as pd import numpy as np data = pd.read_csv('../inputs/puzzle5_input.csv') data = [val for val in data.columns] data[:10]
_____no_output_____
MIT
puzzle_notebooks/puzzle5.ipynb
fromdatavistodatascience/adventofcode2019
Part 5.1 After providing 1 to the only input instruction and passing all the tests, what diagnostic code does the program produce? More Rules: - Opcode 3 takes a single integer as input and saves it to the position given by its only parameter. - Opcode 4 outputs the value of its only parameter. Functions now...
user_ID = 1 numbers = 1002,4,3,4,33 def opcode_instructions(intcode): "Function that breaks the opcode instructions into pieces" str_intcode = str(intcode) opcode = str_intcode[-2:] int_opcode = int(opcode) return int_opcode def extract_p_modes(intcode): "Function that extracts the p_modes" ...
_____no_output_____
MIT
puzzle_notebooks/puzzle5.ipynb
fromdatavistodatascience/adventofcode2019
Selecting the first speech to see what we need to clean.
filename = os.path.join(path, dirs[0]) # dirs is a list, and we are going to study the first element dirs[0] text_file = open(filename, 'r') #open the first file dirs[0] lines = text_file.read() # read the file lines # print what is in the file lines.replace('\n', ' ') # remove the \n symbols by replacing with an empty...
_____no_output_____
MIT
state_of_union_main.ipynb
gequitz/State_of_The_Union_Analysis_NLP
Putting all the speeches into a list, after cleaning them
#The filter() function returns an iterator were the items are filtered #through a function to test if the item is accepted or not. # str.isalpha : checks if it is an alpha character. # lower() : transform everything to lower case # split() : Split a string into a list where each word is a list item # loop over all ...
_____no_output_____
MIT
state_of_union_main.ipynb
gequitz/State_of_The_Union_Analysis_NLP
Count Vectorize
#from notebook #vectorizer = CountVectorizer(stop_words='english') #remove stop words: a, the, and, etc. vectorizer = TfidfVectorizer(stop_words='english', max_df = 0.42, min_df = 0.01) #remove stop words: a, the, and, etc. doc_word = vectorizer.fit_transform(sotu_data) #transform into sparse matrix (0, 1, 2, etc. for...
_____no_output_____
MIT
state_of_union_main.ipynb
gequitz/State_of_The_Union_Analysis_NLP
Compare how similar speeches are to one another
df_similarity = pd.DataFrame(pairwise_similarity.toarray(), index = dirs, columns = dirs) df_similarity.head() #similarity dataframe, compares each document to eachother df_similarity.to_pickle("df_similarity.pkl") #pickle file df_similarity['Speech_str'] = dirs #matrix comparing speech similarity df_similarity['Yea...
_____no_output_____
MIT
state_of_union_main.ipynb
gequitz/State_of_The_Union_Analysis_NLP
Transforming the doc into a dataframe
# We have to convert `.toarray()` because the vectorizer returns a sparse matrix. # For a big corpus, we would skip the dataframe and keep the output sparse. #pd.DataFrame(doc_word.toarray(), index=sotu_data, columns=vectorizer.get_feature_names()).head(10) #doc_word.toarray() makes 7x19 table, otherwise it would be #...
_____no_output_____
MIT
state_of_union_main.ipynb
gequitz/State_of_The_Union_Analysis_NLP
Topic Modeling using nmf
n_topics = 8 # number of topics nmf_model = NMF(n_topics) # create an object doc_topic = nmf_model.fit_transform(doc_word) #break into 10 components like SVD topic_word = pd.DataFrame(nmf_model.components_.round(3), #,"component_9","component_10","component_11","component_12" index = ["component_1","compon...
_____no_output_____
MIT
state_of_union_main.ipynb
gequitz/State_of_The_Union_Analysis_NLP
Top 15 words in each component
n_top_words = 15 feature_names = vectorizer.get_feature_names() print_top_words(nmf_model, feature_names, n_top_words) #Component x Speech H = pd.DataFrame(doc_topic.round(5), index=dirs, #,"component_9","component_10" columns = ["component_1","component_2", "component_3","component_4","compo...
_____no_output_____
MIT
state_of_union_main.ipynb
gequitz/State_of_The_Union_Analysis_NLP
Use NMF to plot top 15 words for each of 8 components def plot_top_words(model, feature_names, n_top_words, title): fig, axes = plt.subplots(2, 4, figsize=(30, 15), sharex=True) axes = axes.flatten() for topic_idx, topic in enumerate(model.components_): top_features_ind = topic.argsort()[:-n_top_words ...
n_top_words = 12 feature_names = vectorizer.get_feature_names() plot_top_words(nmf_model, feature_names, n_top_words, 'Topics in NMF model') #title
_____no_output_____
MIT
state_of_union_main.ipynb
gequitz/State_of_The_Union_Analysis_NLP
Sort speeches Chronologically
H1 = H H1['Speech_str'] = dirs H1['Year'] = H1['Speech_str'].replace('[^0-9]', '', regex=True) H1 = H1.sort_values(by = ['Year']) H1.to_csv("Data_H1.csv", index = False) #Save chronologically sorted speeches in this csv H1.head() H1.to_pickle("H1.pkl") #pickle chronological csv file
_____no_output_____
MIT
state_of_union_main.ipynb
gequitz/State_of_The_Union_Analysis_NLP
Plots of Components over Time (check Powerpoint/Readme for more insights)
plt.subplots(4, 2, figsize=(30, 15), sharex=True) plt.rcParams.update({'font.size': 20}) plt.subplot(4, 2, 1) plt.plot(H1['Year'], H1['component_1'] ) #Label axis and titles for all plots plt.title("19th Century Economic Terms") plt.xlabel("Year") plt.ylabel("Component_1") plt.axhline(y=0.0, color='k', linestyle='...
_____no_output_____
MIT
state_of_union_main.ipynb
gequitz/State_of_The_Union_Analysis_NLP
Component 1: 19th Century Economics
H1.iloc[75:85] #Starts 1831. Peak starts 1868 (apex=1894), Nosedive in 1901 w/ Teddy. 4 Yr resurgence under Taft (1909-1912)
_____no_output_____
MIT
state_of_union_main.ipynb
gequitz/State_of_The_Union_Analysis_NLP
Component 2: Modern Economic Language
H1.iloc[205:215] #1960s: Starts under JFK in 1961, peaks w/ Clinton, dips post 9/11 Bush, resurgence under Obama
_____no_output_____
MIT
state_of_union_main.ipynb
gequitz/State_of_The_Union_Analysis_NLP
Component 3: Growth of US Government and Federal Programs
H1.iloc[155:165] #1921, 1929-1935. Big peak in 1946-1950 (1951 Cold War). 1954-1961 Eisenhower. Low after Reagan Revolution (1984)
_____no_output_____
MIT
state_of_union_main.ipynb
gequitz/State_of_The_Union_Analysis_NLP
Component 4: Early Foreign Policy and War
H1.iloc[30:40] #Highest from 1790-1830, Washington to Jackson
_____no_output_____
MIT
state_of_union_main.ipynb
gequitz/State_of_The_Union_Analysis_NLP
Component 5: Progressive Era, Roaring 20s
H1.iloc[115:125] #Peaks in 1900-1930.Especially Teddy Roosevelt. Dip around WW1
_____no_output_____
MIT
state_of_union_main.ipynb
gequitz/State_of_The_Union_Analysis_NLP
Component 6: War Before, During, and After the Civil War
H1.iloc[70:80] #Starts w/ Jackson 1829, Peaks w/ Mexican-American War (1846-1848). Drops 60% w/ Lincoln. Peak ends w/ Johnson 1868. Remains pretty low after 1876 (Reconstruction ends)
_____no_output_____
MIT
state_of_union_main.ipynb
gequitz/State_of_The_Union_Analysis_NLP
Component 7: World Wars and Korean War
H1.iloc[155:165] #Minor Peak around WW1. Masssive spike a response of Cold War, Korean War (1951). Eisenhower drops (except 1960 U2). Johnson Vietnam. Peaks again 1980 (Jimmy Carter foreign policy crises)
_____no_output_____
MIT
state_of_union_main.ipynb
gequitz/State_of_The_Union_Analysis_NLP
Component 8: Iraq War and Terrorism
H1.iloc[210:220] #Minor peak w/ Bush 1990. BIG peak w/ Bush 2002. Ends w/ Obama 2009. Resurgence in 2016/18 (ISIS?)
_____no_output_____
MIT
state_of_union_main.ipynb
gequitz/State_of_The_Union_Analysis_NLP
Word Cloud
from wordcloud import WordCloud, STOPWORDS, ImageColorGenerator speech_name = 'Lincoln_1864.txt' sotu_dict[path + '\\' + speech_name] #example = sotu_data[0] example = sotu_dict[path + '\\' + speech_name] wordcloud = WordCloud(max_words=100).generate(example) plt.title("WordCloud of " + speech_name) plt.imshow(wordclou...
_____no_output_____
MIT
state_of_union_main.ipynb
gequitz/State_of_The_Union_Analysis_NLP
Average Monthly Temperatures, 1970-2004**Date:** 2021-12-02**Reference:**
library(TTR) options( jupyter.plot_mimetypes = "image/svg+xml", repr.plot.width = 7, repr.plot.height = 5 )
_____no_output_____
MIT
jupyter/2_time_series/2_03_ljk_decompose_seasonal.ipynb
ljk233/R249
SummaryThe aim of this notebook was to show how to decompose seasonal time series data using **R** so the trend, seasonal and irregular components can be estimated.Data on the average monthly temperatures in central England from January 1970 to December 2004 was plotted.The series was decomposed using the `decompose` ...
monthlytemps <- read.csv("..\\..\\data\\moderntemps.csv") head(monthlytemps) modtemps <- monthlytemps$temperature
_____no_output_____
MIT
jupyter/2_time_series/2_03_ljk_decompose_seasonal.ipynb
ljk233/R249
Plot the time series
ts_modtemps <- ts(modtemps, start = c(1970, 1), frequency = 12) plot.ts(ts_modtemps, xlab = "year", ylab = "temperature")
_____no_output_____
MIT
jupyter/2_time_series/2_03_ljk_decompose_seasonal.ipynb
ljk233/R249
The time series is highly seasonal with little evidence of a trend.There appears to be a constant level of approximately 10$^{\circ}$C. Decompose the dataUse the `decompose` function from `R.stats` to return estimates of the trend, seasonal, and irregular components of the time series.
decomp_ts <- decompose(ts_modtemps)
_____no_output_____
MIT
jupyter/2_time_series/2_03_ljk_decompose_seasonal.ipynb
ljk233/R249
Seasonal factorsCalculate the seasonal factors of the decomposed time series.Cast the `seasonal` time series object held in `decomp_ts` to a `vector`, slice the new vector to isolate a single period, and then cast the sliced vector to a named `matrix`.
sf <- as.vector(decomp_ts$seasonal) (matrix(sf[1:12], dimnames = list(month.abb, c("factors"))))
_____no_output_____
MIT
jupyter/2_time_series/2_03_ljk_decompose_seasonal.ipynb
ljk233/R249
_Add a comment_ Plot the componentsPlot the trend, seasonal, and irregular components in a single graphic.
plot(decomp_ts, xlab = "year")
_____no_output_____
MIT
jupyter/2_time_series/2_03_ljk_decompose_seasonal.ipynb
ljk233/R249
Plot the individual components of the decomposition by accessing the variables held in the `tsdecomp`.This will generally make the components easier to understand.
plot(decomp_ts$trend, xlab = "year", ylab = "temperature (Celsius)") title(main = "Trend component") plot(decomp_ts$seasonal, xlab = "year", ylab = "temperature (Celsius)") title(main = "Seasonal component") plot(decomp_ts$random, xlab = "year", ylab = "temperature (Celsius)") title(main = "Irregular component")
_____no_output_____
MIT
jupyter/2_time_series/2_03_ljk_decompose_seasonal.ipynb
ljk233/R249
_Add comment on trend, seasonal, and irregular components.__Which component dominates the series?_ Seasonal adjusted plotPlot the seasonally adjusted series by subtracting the seasonal factors from the original series.
adjusted_ts <- ts_modtemps - decomp_ts$seasonal plot(adjusted_ts, xlab = "year", ylab = "temperature (Celsius)") title(main = "Seasonally adjusted series")
_____no_output_____
MIT
jupyter/2_time_series/2_03_ljk_decompose_seasonal.ipynb
ljk233/R249
This new seasonally adjusted series only contains the trend and irregular components, so it can be treated as if it is non-seasonal data.Estimate the trend component by taking the simple moving order of order 35.
sma35_adjusted_ts <- SMA(adjusted_ts, n = 35) plot.ts(sma35_adjusted_ts, xlab = "year", ylab = "temperature (Celsius)") title(main = "Trend component (ma35)")
_____no_output_____
MIT
jupyter/2_time_series/2_03_ljk_decompose_seasonal.ipynb
ljk233/R249
PySchools without Thomas High School 9th graders Dependencies and data
# Dependencies import os import numpy as np import pandas as pd # School data school_path = os.path.join('data', 'schools.csv') # school data path school_df = pd.read_csv(school_path) # Student data student_path = os.path.join('data', 'students.csv') # student data path student_df = pd.read_csv(student_path) school_d...
_____no_output_____
MIT
pyschools/analysis2.ipynb
tri-bui/sandbox-analytics
Clean student names
# Prefixes to remove: "Miss ", "Dr. ", "Mr. ", "Ms. ", "Mrs. " # Suffixes to remove: " MD", " DDS", " DVM", " PhD" fixes_to_remove = ['Miss ', '\w+\. ', ' [DMP]\w?[DMS]'] # regex for prefixes and suffixes str_to_remove = r'|'.join(fixes_to_remove) # join into a single raw str # Remove inappropriate prefixes and suffix...
['Juan', 'Noah', 'Cory', 'Omar', 'Eric', 'Ryan', 'Sean', 'Jon', 'Cody', 'Todd', 'Erik', 'Greg', 'Adam', 'Seth', 'Tony', 'Mark'] ['V', 'IV', 'Jr.', 'III', 'II']
MIT
pyschools/analysis2.ipynb
tri-bui/sandbox-analytics
Merge data
# Add binary vars for passing score student_df['pass_read'] = (student_df.reading_score >= 70).astype(int) # passing reading score student_df['pass_math'] = (student_df.math_score >= 70).astype(int) # passing math score student_df['pass_both'] = np.min([student_df.pass_read, student_df.pass_math], axis=0) # passing bot...
<class 'pandas.core.frame.DataFrame'> Int64Index: 39170 entries, 0 to 39169 Data columns (total 17 columns): # Column Non-Null Count Dtype --- ------ -------------- ----- 0 Student ID 39170 non-null int64 1 student_name 39170 non-null object 2 gender...
MIT
pyschools/analysis2.ipynb
tri-bui/sandbox-analytics
District summary
# District summary district_summary = pd.DataFrame(school_df[['size', 'budget']].sum(), columns=['District']).T district_summary['Total Schools'] = school_df.shape[0] district_summary = district_summary[['Total Schools', 'size', 'budget']] district_summary_cols = ['Total Schools', 'Total Students', 'Total Budget'] dist...
_____no_output_____
MIT
pyschools/analysis2.ipynb
tri-bui/sandbox-analytics
School summary
# School cols school_cols = ['type', 'size', 'budget', 'budget_per_student', 'reading_score', 'math_score', 'pass_read', 'pass_math', 'pass_both'] school_cols_new = ['School Type', 'Total Students', 'Total Budget', 'Budget Per Student'] school_cols_new += score_cols_new # School summary school_summary ...
_____no_output_____
MIT
pyschools/analysis2.ipynb
tri-bui/sandbox-analytics
Scores by grade
# Reading scores by grade of each school grade_read_scores = pd.pivot_table(df, index='school_name', columns='grade', values='reading_score', aggfunc='mean').round(2) grade_read_scores.index.name = None grade_read_scores.columns.name = 'Reading scores' grade_read_scores = grade_read_...
_____no_output_____
MIT
pyschools/analysis2.ipynb
tri-bui/sandbox-analytics
Scores by budget per student
# Scores by spending spending_scores = df.groupby('spending_lvl')[score_cols].mean().round(2) for col in spending_scores.columns: if "pass" in col: spending_scores[col] = (spending_scores[col] * 100).astype(int) spending_scores # Formatting spending_scores.index.name = 'Spending Level' spending_scores.colum...
_____no_output_____
MIT
pyschools/analysis2.ipynb
tri-bui/sandbox-analytics
Scores by school size
# Scores by school size size_scores = df.groupby('school_size')[score_cols].mean().round(2) for col in size_scores.columns: if "pass" in col: size_scores[col] = (size_scores[col] * 100).astype(int) size_scores # Formatting size_scores.index.name = 'School Size' size_scores.columns = score_cols_new size_scor...
_____no_output_____
MIT
pyschools/analysis2.ipynb
tri-bui/sandbox-analytics
Scores by school type
# Scores by school type type_scores = df.groupby('type')[score_cols].mean().round(2) for col in type_scores.columns: if "pass" in col: type_scores[col] = (type_scores[col] * 100).astype(int) type_scores # Formatting type_scores.index.name = 'School Type' type_scores.columns = score_cols_new type_scores
_____no_output_____
MIT
pyschools/analysis2.ipynb
tri-bui/sandbox-analytics
InstructionsImplement multi output cross entropy loss in pytorch.Throughout this whole problem we use multioutput models:* predicting 4 localization coordinates* predicting 4 keypoint coordinates + whale id and callosity pattern* predicting whale id and callosity patternIn order for that to work your loss function nee...
def solution(outputs, targets): """ Args: outputs: list of torch.autograd.Variables containing model outputs targets: list of torch.autograd.Variables containing targets for each output Returns: loss_value: torch.autograd.Variable object """ return loss_value
_____no_output_____
MIT
resources/whales/tasks/task5.ipynb
pknut/minerva
**ANALYSIS OF FINANCIAL INCLUSION IN EAST AFRICA BETWEEN 2016 TO 2018** DEFINING QUESTION The research problem is to figure out how we can predict which individuals are most likely to have or use a bank account. METRIC FOR SUCCESS My solution procedure will be to help provide an indication of the state of financial ...
# importing libraries import pandas as pd import numpy as np import seaborn as sns import matplotlib.pyplot as plt
_____no_output_____
MIT
Moringa_Data_Science_Core_W2_Independent_Project_2021_09_Moreen_Mugambi_Python_Notebook_ipyn.ipynb
MoreenMarutaData/FINANCIAL-INCLUSION-IN-EAST-AFRICA-MORINGA-CORE-WEEK-2-PROJECT
READING AND CHECKING DATA
# loading and viewing variable definitions dataset url = "http://bit.ly/VariableDefinitions" vb_df = pd.read_csv(url) vb_df # loading and viewing financial dataset url2 = "http://bit.ly/FinancialDataset" fds = pd.read_csv(url2) fds fds.shape fds.head() fds.tail() fds.dtypes fds.columns fds.info() fds.describe() fds.d...
_____no_output_____
MIT
Moringa_Data_Science_Core_W2_Independent_Project_2021_09_Moreen_Mugambi_Python_Notebook_ipyn.ipynb
MoreenMarutaData/FINANCIAL-INCLUSION-IN-EAST-AFRICA-MORINGA-CORE-WEEK-2-PROJECT
EXTERNAL DATA SOURCE VALIDATION FinAccess Kenya 2018: https://fsdkenya.org/publication/finaccess2019/Finscope Rwanda 2016: http://www.statistics.gov.rw/publication/finscope-rwanda-2016 Finscope Tanzania 2017: http://www.fsdt.or.tz/finscope/Finscope Uganda 2018: http://fsduganda.or.ug/finscope-2018-survey-report/ CLEA...
fds.head(2) # CHECKING FOR OUTLIERS IN YEAR COLUMN sns.boxplot(x=fds['year']) fds.shape # dropping year column outliers fds1= fds[fds['year']<2020] fds1.shape # CHECKING FOR OUTLIERS IN HOUSEHOLD SIZE COLUMN sns.boxplot(x=fds1['household_size']) # dropping household size outliers fds2 =fds1[fds1['ho...
_____no_output_____
MIT
Moringa_Data_Science_Core_W2_Independent_Project_2021_09_Moreen_Mugambi_Python_Notebook_ipyn.ipynb
MoreenMarutaData/FINANCIAL-INCLUSION-IN-EAST-AFRICA-MORINGA-CORE-WEEK-2-PROJECT
**EXPLORATORY ANALYSIS** 1.UNIVARIATE ANALYSIS a. NUMERICAL VARIABLES MODE
fds5['year'].mode() fds5['household_size'].mode() fds5['age_of_respondent'].mode()
_____no_output_____
MIT
Moringa_Data_Science_Core_W2_Independent_Project_2021_09_Moreen_Mugambi_Python_Notebook_ipyn.ipynb
MoreenMarutaData/FINANCIAL-INCLUSION-IN-EAST-AFRICA-MORINGA-CORE-WEEK-2-PROJECT
MEAN
fds5['age_of_respondent'].mean() fds5['household_size'].mean() fds5.mean()
_____no_output_____
MIT
Moringa_Data_Science_Core_W2_Independent_Project_2021_09_Moreen_Mugambi_Python_Notebook_ipyn.ipynb
MoreenMarutaData/FINANCIAL-INCLUSION-IN-EAST-AFRICA-MORINGA-CORE-WEEK-2-PROJECT
MEDIAN
fds5['age_of_respondent'].median() fds5['household_size'].median() fds5.median()
_____no_output_____
MIT
Moringa_Data_Science_Core_W2_Independent_Project_2021_09_Moreen_Mugambi_Python_Notebook_ipyn.ipynb
MoreenMarutaData/FINANCIAL-INCLUSION-IN-EAST-AFRICA-MORINGA-CORE-WEEK-2-PROJECT
RANGE
a = fds5['age_of_respondent'].max() b = fds5['age_of_respondent'].min() c = a-b print('The range of the age for the respondents is', c) d = fds5['household_size'].max() e = fds5['household_size'].min() f = d-e print('The range of the household_sizes is', f)
The range of the household_sizes is 8.0
MIT
Moringa_Data_Science_Core_W2_Independent_Project_2021_09_Moreen_Mugambi_Python_Notebook_ipyn.ipynb
MoreenMarutaData/FINANCIAL-INCLUSION-IN-EAST-AFRICA-MORINGA-CORE-WEEK-2-PROJECT
QUANTILE AND INTERQUANTILE
fds5.quantile([0.25,0.5,0.75]) # FINDING THE INTERQUANTILE RANGE = IQR Q3 = fds5['age_of_respondent'].quantile(0.75) Q2 = fds5['age_of_respondent'].quantile(0.25) IQR= Q3-Q2 print('The IQR for the respondents age is', IQR) q3 = fds5['household_size'].quantile(0.75) q2 = fds5['household_size'].quantile(0.25) iqr = q3-q2...
_____no_output_____
MIT
Moringa_Data_Science_Core_W2_Independent_Project_2021_09_Moreen_Mugambi_Python_Notebook_ipyn.ipynb
MoreenMarutaData/FINANCIAL-INCLUSION-IN-EAST-AFRICA-MORINGA-CORE-WEEK-2-PROJECT
STANDARD DEVIATION
fds5.std()
_____no_output_____
MIT
Moringa_Data_Science_Core_W2_Independent_Project_2021_09_Moreen_Mugambi_Python_Notebook_ipyn.ipynb
MoreenMarutaData/FINANCIAL-INCLUSION-IN-EAST-AFRICA-MORINGA-CORE-WEEK-2-PROJECT
VARIANCE
fds5.var()
_____no_output_____
MIT
Moringa_Data_Science_Core_W2_Independent_Project_2021_09_Moreen_Mugambi_Python_Notebook_ipyn.ipynb
MoreenMarutaData/FINANCIAL-INCLUSION-IN-EAST-AFRICA-MORINGA-CORE-WEEK-2-PROJECT
KURTOSIS
fds5.kurt()
_____no_output_____
MIT
Moringa_Data_Science_Core_W2_Independent_Project_2021_09_Moreen_Mugambi_Python_Notebook_ipyn.ipynb
MoreenMarutaData/FINANCIAL-INCLUSION-IN-EAST-AFRICA-MORINGA-CORE-WEEK-2-PROJECT
SKEWNESS
fds5.skew()
_____no_output_____
MIT
Moringa_Data_Science_Core_W2_Independent_Project_2021_09_Moreen_Mugambi_Python_Notebook_ipyn.ipynb
MoreenMarutaData/FINANCIAL-INCLUSION-IN-EAST-AFRICA-MORINGA-CORE-WEEK-2-PROJECT
b. CATEGORICAL MODE
fds5.mode().head(1) fds5['age_of_respondent'].plot(kind="hist") plt.xlabel('ages of respondents') plt.ylabel('frequency') plt.title(' Frequency of the ages of the respondents') country=fds5['country'].value_counts() print(country) # Plotting the pie chart colors=['pink','white','cyan','yellow'] country.plot(kind='pie'...
_____no_output_____
MIT
Moringa_Data_Science_Core_W2_Independent_Project_2021_09_Moreen_Mugambi_Python_Notebook_ipyn.ipynb
MoreenMarutaData/FINANCIAL-INCLUSION-IN-EAST-AFRICA-MORINGA-CORE-WEEK-2-PROJECT
CONCLUSION AND RECOMMENDATION Most of the data was collected in Rwanda.Most of the data was collected in Rural areas.Most of those who were interviewed were women.Most of the population has mobile phones.There were several outliers.Since 75% of the population has phones, phones should be used as the main channel for i...
fds5.head() #@title Since i am predicting the likelihood of the respondents using the bank,I shall be comparing all variables against the bank account column.
_____no_output_____
MIT
Moringa_Data_Science_Core_W2_Independent_Project_2021_09_Moreen_Mugambi_Python_Notebook_ipyn.ipynb
MoreenMarutaData/FINANCIAL-INCLUSION-IN-EAST-AFRICA-MORINGA-CORE-WEEK-2-PROJECT
NUMERICAL VS NUMERICAL
sns.pairplot(fds5) plt.show() # pearson correlation of numerical variables sns.heatmap(fds5.corr(),annot=True) plt.show() # possible weak correlation fds5.corr()
_____no_output_____
MIT
Moringa_Data_Science_Core_W2_Independent_Project_2021_09_Moreen_Mugambi_Python_Notebook_ipyn.ipynb
MoreenMarutaData/FINANCIAL-INCLUSION-IN-EAST-AFRICA-MORINGA-CORE-WEEK-2-PROJECT
CATEGORICAL VS CATEGORICAL
# Grouping bank usage by country country1 = fds5.groupby('country')['bank account'].value_counts(normalize=True).unstack() colors= ['lightpink', 'skyblue'] country1.plot(kind='bar', figsize=(8, 6), color=colors, stacked=True) plt.title('Bank usage by country', fontsize=15, y=1.015) plt.xlabel('country', fontsize=14, ...
_____no_output_____
MIT
Moringa_Data_Science_Core_W2_Independent_Project_2021_09_Moreen_Mugambi_Python_Notebook_ipyn.ipynb
MoreenMarutaData/FINANCIAL-INCLUSION-IN-EAST-AFRICA-MORINGA-CORE-WEEK-2-PROJECT
NUMERICAL VS CATEGORICAL IMPLEMENTING AND CHALLENGING SOLUTION
_____no_output_____
MIT
Moringa_Data_Science_Core_W2_Independent_Project_2021_09_Moreen_Mugambi_Python_Notebook_ipyn.ipynb
MoreenMarutaData/FINANCIAL-INCLUSION-IN-EAST-AFRICA-MORINGA-CORE-WEEK-2-PROJECT
Most of those interviewed do not have bank accounts of which 80% is the uneducated.Most of the population that participated is married,followed by single/never married.Most of the population has primary school education level.Most of the population is involved in farming followed by self employment.Bank usage has more ...
# Multivariate analysis - This is a statistical analysis that involves observation and analysis of more than one statistical outcome variable at a time # LETS MAKE A COPY fds_new = fds5.copy() fds_new.columns fds_new.dtypes # IMPORTING THE LABEL ENCODER from sklearn.preprocessing import LabelEncoder le = LabelEncoder...
_____no_output_____
MIT
Moringa_Data_Science_Core_W2_Independent_Project_2021_09_Moreen_Mugambi_Python_Notebook_ipyn.ipynb
MoreenMarutaData/FINANCIAL-INCLUSION-IN-EAST-AFRICA-MORINGA-CORE-WEEK-2-PROJECT
FACTOR ANALYSIS
# Installing factor analyzer !pip install factor_analyzer==0.2.3 from factor_analyzer.factor_analyzer import calculate_bartlett_sphericity chi_square_value,p_value=calculate_bartlett_sphericity(fds_new) chi_square_value, p_value # In Bartlett ’s test, the p-value is 0. The test was statistically significant, # in...
_____no_output_____
MIT
Moringa_Data_Science_Core_W2_Independent_Project_2021_09_Moreen_Mugambi_Python_Notebook_ipyn.ipynb
MoreenMarutaData/FINANCIAL-INCLUSION-IN-EAST-AFRICA-MORINGA-CORE-WEEK-2-PROJECT
Configure
sample_size = 0 max_closure_size = 10000 max_distance = 0.22 cluster_distance_threshold = 0.155 super_cluster_distance_threshold = 0.205 num_candidates = 1000 eps = 0.000001 model_filename = '../data/models/anc-triplet-bilstm-100-512-40-05.pth' # process_nicknames = True # werelate_names_filename = 'givenname_similar_...
_____no_output_____
MIT
reports/80_cluster_anc_triplet-initial.ipynb
rootsdev/nama
Read WeRelate names into all_namesLater, we'll want to read frequent FS names into all_names
# TODO rewrite this in just a few lines using pandas def load_werelate_names(path, is_surname): name_variants = defaultdict(set) with fopen(path, mode="r", encoding="utf-8") as f: is_header = True for line in f: if is_header: is_header = False continue...
_____no_output_____
MIT
reports/80_cluster_anc_triplet-initial.ipynb
rootsdev/nama
Read nicknames and remove from names
def load_nicknames(path): nicknames = defaultdict(set) with fopen(path, mode="r", encoding="utf-8") as f: for line in f: names = line.rstrip().split(" ") # normalize should only return a single name piece, but loop just in case for name_piece in normalize(names[0], Fa...
_____no_output_____
MIT
reports/80_cluster_anc_triplet-initial.ipynb
rootsdev/nama
Map names to ids
def map_names_to_ids(names): ids = range(len(names)) return dict(zip(names, ids)), dict(zip(ids, names)) name_ids, id_names = map_names_to_ids(all_names) print(next(iter(name_ids.items())), next(iter(id_names.items())))
_____no_output_____
MIT
reports/80_cluster_anc_triplet-initial.ipynb
rootsdev/nama