First time using TabPy and have the connections successfully set up. Within the "Create Calculated Field" button in Tableau, I have tried
SCRIPT_REAL("
import numpy as np
import pandas as pd
")
which results in "SCRIPT_REAL is being called with (string), did you mean (string, ...)?
Additionally, how do I refer to the dataset such as I did in Python to execute the following?
data = pd.read_csv("C:/Users/.../dataset.csv")
data.head()
plt.figure(figsize=(7,7))
plt.pie(data['stroke'].value_counts(sort = True),
explode = (0.05, 0),
labels = data['stroke'].value_counts(sort = True).index,
colors = ["blue","green"],
autopct = '%1.1f%%')
plt.title('Pie Chart')
plt.show()
Related
I have a csv of relevant points with latitude and longitude and trying to get the nearest
building data to each point and add a column to the csv (or panda) in python. Tried using Pyrosm and various libraries but can't seem to prune the data to get the nearest building and then add the data. Thanks
This is what I have
from pyrosm import OSM
from pyrosm import get_data
import geopandas as gpd
from sklearn.neighbors import BallTree
import numpy as np
import osmnx as ox
# get rid of weird error
import shapely
import warnings
from shapely.errors import ShapelyDeprecationWarning
import csv
def get_gig_data(csv_fname):
with open(csv_fname, "r", encoding="latin-1") as gig_records:
for gig_record in csv.reader(gig_records):
yield gig_record
def main():
warnings.filterwarnings("ignore", category=ShapelyDeprecationWarning)
chicago_osm = OSM(get_data("chicago"))
#get a Point of Interest GeoDataFrame
points_of_interest = chicago_osm.get_pois() #can use a custom filter if we want to filter the types, but I think no filter might be the best
# get buildings nodes and edges
nodes, edges = chicago_osm.get_network(nodes=True, network_type="walking")
buildings = chicago_osm.get_buildings()
b_cnt = len(buildings)
G = chicago_osm.to_graph(nodes, edges)
#nodes = get_igraph_nodes(G)
buildings['geometry'] = buildings.centroid
# poi_list = np.asarray([ point.coords for point in points_of_interest['geometry'] ]) #if point.geom_type == point])
#print(poi_list.shape)
#tree = BallTree( np.asarray([ point.coords for point in points_of_interest['geometry'] if point.geom_type == point]), metric="manhattan") #Note: the scipy implementation of manhattan/cityblock distance might be faster according to the internet bc it uses a C function
#Read in the gig work data - I think the best way to do this will probably be with the CSV.reader with open thing because it will go line by line and save a ton of memory
'''for i in points_of_interest:
print('Type: ', type(i) , ' ',i)'''
gig_fp = "data_sample.csv"
#gig_data = gpd.read_file(gig_fp)
iter_gig = iter(get_gig_data(gig_fp))
next(iter_gig)
ids=dict()
for building in buildings.iterrows():
#print(type(building[1][32]) , ' ', building[1][32])
#tup = tuple(float(x) for x in [trip[17][8:-1].split()])
ids[building[1][32]] = building
#make the tree that determines closest POI
#if we use the CSV reader this for loop will be done already
for trip in iter_gig:
# Using generator so this should be efficient memory wise.
tup = tuple([float(x) for x in trip[17][8:-1].split() ])
print(type(tup), ' ', tup)
src_ids,euclidean_distance=ox.distance.nearest_nodes(G,tup)
src_ids, euclidean_distance= ox.distance.nearest_nodes(G,tup)
# find nearest node
#THEN ADD THE PICKUP AND DROPOFF IDS TO THIS TUPLE AND ADD TO A NEW NP ARRAY
if __name__ == '__main__':
main()
I'm trying to implement KMeans by PySpark feeding with a ndarray of dimensions: (289, 768). But when call KMEANS.fit, an error occurs:
text = np.array(df_low_conf.select("text").collect()).reshape(-1)
model = SentenceTransformer('neuralmind/bert-base-english-cased')
sentence_embeddings = model.encode(text)
kmeans = KMeans(k = num_clusters, initMode = 'k-means||', maxIter= 2000, initSteps = 10)
model = kmeans.fit(sentence_embeddings)
'numpy.ndarray' object has no attribute '_jdf'
Is it possible in PySpark? Because I tried on pandas and it was fine. If you have any advice or tip, please send me a message.
I am running the code through jupyter on EMR, pyspark version 3.3.0.
I have two dataframes that I have preprocessed with the pyspark.ml.feature functions (OneHotEncoder, StringIndexer, VectorAssembler). The first dataframe, lets call it df_good, has 5 features, the second dataframe, lets call it df_bad, omits 2 of the features from df_good. The underlying dataset used to generate the two datasets is the same, the code to generate the datasets is identical (other than two features not be included in the VectorAssembler inputCols for df_bad).
Below is the code I am using to train the model:
import pyspark.sql.functions as F
from pyspark.sql.types import ArrayType, DoubleType
from pyspark.ml.classification import LogisticRegression
def split_array(col):
def to_list(v):
return v.toArray().tolist()
return F.udf(to_list, ArrayType(DoubleType()))(col)
def train_model(df):
train_df = df.selectExpr("label as label", "features as features")
logit = LogisticRegression()
logit = logit.setFamily("multinomial")
logit_mod = logit.fit(train_df)
df = logit_mod.transform(df)
df = df.withColumn("pred", split_array(F.col("probability"))[0])
return df
Here is where things get weird.
If I run the code below it works and runs in 10-20 seconds each:
df_good = spark.read.parquet("<s3_location_good>")
df_good = train_model(df_good)
df_good.select(F.sum("pred")).show()
df_bad = spark.read.parquet("<s3_location_bad>")
df_bad = train_model(df_bad)
df_bad.select(F.sum("pred")).show()
If I change the order, the code completely hangs on df_bad:
df_bad = spark.read.parquet("<s3_location_bad>")
df_bad = train_model(df_bad)
df_bad.select(F.sum("pred")).show()
df_good = spark.read.parquet("<s3_location_good>")
df_good = train_model(df_good)
df_good.select(F.sum("pred")).show()
The data is unchanged, the code is the same, the behavior is always the same.
Any thoughts are appreciated.
I have created a function for applying OLS regression and just getting the model parameters. I used groupby and applyInPandas but it's taking too much of time. Is there are more efficient way to work around this?
Note: I din't had to use groupby as all features have many levels but as I cannot use applyInPandas without it so I created a dummy feature as 'group' having the same value as 1.
Code
import pandas as pd
import statsmodels.api as sm
from pyspark.sql.functions import lit
pdf = pd.DataFrame({
'x':[3,6,2,0,1,5,2,3,4,5],
'y':[0,1,2,0,1,5,2,3,4,5],
'z':[2,1,0,0,0.5,2.5,3,4,5,6]})
df = sqlContext.createDataFrame(pdf)
result_schema =StructType([
StructField('index',StringType()),
StructField('coef',DoubleType())
])
def ols(pdf):
y_column = ['z']
x_column = ['x', 'y']
y = pdf[y_column]
X = pdf[x_column]
model = sm.OLS(y, X).fit()
param_table =pd.DataFrame(model.params, columns = ['coef']).reset_index()
return param_table
#adding a new column to apply groupby
df = df.withColumn('group', lit(1))
#applying function
data = df.groupby('group').applyInPandas(ols, schema = result_schema)
Final output sample
index coef
x 0.183246073
y 0.770680628
In MongoDB, I have a database named twitter. From the database, I was trying to extract 5 languages that were used more to tweet. And I wanted to show the result using a chart. For this, I used pandas and matplotlib. But while executing the code, it says, TypeError: Empty 'DataFrame': no numeric data to plot
Here is my code:
import pymongo
import pandas as pd
import matplotlib.pyplot as plt
#conect with MongoClient daemon mongod
client_con = pymongo.MongoClient()
#conect with db
mydb = client_con["twitter"]
#conect with collection
mycol = mydb["twitterCol"]
#print the records
print mycol.count()
#intiate pandas
tweets = pd.DataFrame()
tweets_data = []
tweets_data= mydb.mycol.find()
"""Next, 3 columns will be added to the tweets
DataFrame called text, lang, and country.
Text column contains the tweet,
Lang column contains the language in which the
tweet was written,
and Country the country from which the tweet was sent.
"""
tweets['text'] = map(lambda tweet: tweet["text"], tweets_data)
tweets['lang'] = map(lambda tweet: tweet['lang'], tweets_data)
tweets['country'] = map(lambda tweet: tweet['place']['country'] if tweet['place'] != None else None, tweets_data)
tweets_by_lang = tweets['lang'].value_counts()
tweets_by_lang = tweets_by_lang.astype(int)
fig, ax = plt.subplots()
ax.tick_params(axis='x', labelsize=15)
ax.tick_params(axis='y', labelsize=10)
ax.set_xlabel('Languages', fontsize=15)
ax.set_ylabel('Number of tweets' , fontsize=15)
ax.set_title('Top 5 languages', fontsize=15, fontweight='bold')
tweets_by_lang.plot(ax=ax, kind='bar', color='red')
How to resolve this problem?