-
Notifications
You must be signed in to change notification settings - Fork 0
/
app.py
136 lines (106 loc) · 6.32 KB
/
app.py
1
2
3
4
5
6
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
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
from flask import Flask, render_template, request, redirect, Response, jsonify
import pandas as pd
from sklearn.decomposition import PCA
# from pca_component import remove_extra_cols, pca, task2_top3attr, task3_get_MDS_data, get_top3_pca_atttr_data
import numpy as np
import json
app = Flask(__name__)
# Opening JSON file
f = open('datasets/nyc.geo.json',)
geo_data = json.load(f)
stratified_sample = pd.read_csv('datasets/stratified_sampled_data.csv')
grouped_multiple_data = pd.read_csv('datasets/grouped_multiple_data.csv')
data_for_pca = stratified_sample[["review_scores_rating_zscore","price_zscore","crime_zscore"]]
def do_PCA(data):
# PCA implementation
covar_matrix = PCA(n_components = len(data.columns))
X = pd.DataFrame(covar_matrix.fit_transform(data))
# Explained variance % per PC
variance = np.round(covar_matrix.explained_variance_ratio_*100,decimals=2)
# Cumulative explained variance % per PC
cumulative_var_percentage = np.round(np.cumsum(np.round(variance,decimals=2)),decimals=2)
# Get PC1 and PC2 values to plot ahead
trans_matrix = X[[0,1]]
return variance,cumulative_var_percentage,trans_matrix
variance,cumulative_var_percentage,trans_matrix = do_PCA(data_for_pca)
trans_matrix["neighbourhood_group"] = stratified_sample["neighbourhood_group"]
trans_matrix["index"] = stratified_sample["index"]
trans_matrix.columns = ["PC1","PC2","neighbourhood_group","index"]
trans_matrix = trans_matrix.reset_index()
@app.route("/")
def index():
return render_template("index.html")
@app.route("/geodata")
def get_geodata():
return geo_data
@app.route("/map")
def get_map_data():
map_data = stratified_sample[["index","neighbourhood_group","longitude", "latitude"]]
map_data = map_data.to_json(orient = "records")
return map_data
@app.route("/pricescatter")
def get_pricescatter_data():
pricescatter_data = stratified_sample[["index","neighbourhood_group","price", "review_scores_rating"]]
pricescatter_data = pricescatter_data.to_json(orient = "records")
return pricescatter_data
@app.route("/pcascatter")
def get_pcascatter_data():
pcascatter_data = trans_matrix[["index","neighbourhood_group","PC1", "PC2"]]
pcascatter_data = pcascatter_data.to_json(orient = "records")
return pcascatter_data
@app.route("/barcrime/<indexids>",methods = ['GET'])
def get_barcrime_data(indexids):
json_indexids = json.dumps(indexids)
month_dict = {1:"JAN",2:"FEB",3:"MAR",4:"APR",5:"MAY",6:"JUN",7:"JUL",8:"AUG",9:"SEP",10:"OCT",11:"NOV",12:"DEC"}
if json_indexids == "\"all\"":
brushed_zipcodes = grouped_multiple_data.zipcodes.unique()
grouped_filtered = grouped_multiple_data.loc[grouped_multiple_data['zipcodes'].isin(brushed_zipcodes)]
else:
json_indexes = json.loads(json_indexids)[1:-1]
json_indexes = list(map(int,json_indexes.split(',')))
stratified_sample_new = stratified_sample.set_index('index')
brushed_sample = stratified_sample_new[stratified_sample_new.index.isin(json_indexes)]["zipcodes"].unique().astype(np.int32)
# print(stratified_sample[stratified_sample.index.isin(json_indexes)][["neighbourhood_group","name"]])
grouped_filtered = grouped_multiple_data.loc[grouped_multiple_data['zipcodes'].isin(brushed_sample)]
grouped_by_month = grouped_filtered[["CMPLNT_MONTH","CMPLNT_NUM_count"]].groupby("CMPLNT_MONTH").sum().reset_index()
grouped_by_month = grouped_by_month[grouped_by_month["CMPLNT_MONTH"]>6]
grouped_by_month["group"] = grouped_by_month["CMPLNT_MONTH"].map(month_dict)
grouped_by_month["value"] = grouped_by_month["CMPLNT_NUM_count"]
return grouped_by_month[["group","value"]].to_json(orient="records")
@app.route("/barcrimefilter/<boroughs>",methods = ['GET'])
def get_barcrime_data_filter(boroughs):
boroughs = boroughs.replace("\"","")
borough_list = json.dumps(boroughs)
borough_list = json.loads(borough_list)[1:-1]
borough_list = list(borough_list.split(','))
filtered_sample = stratified_sample.loc[stratified_sample["neighbourhood_group"].isin(borough_list)]["zipcodes"].unique().astype(np.int32)
month_dict = {1:"JAN",2:"FEB",3:"MAR",4:"APR",5:"MAY",6:"JUN",7:"JUL",8:"AUG",9:"SEP",10:"OCT",11:"NOV",12:"DEC"}
grouped_filtered = grouped_multiple_data.loc[grouped_multiple_data['zipcodes'].isin(filtered_sample)]
grouped_by_month = grouped_filtered[["CMPLNT_MONTH","CMPLNT_NUM_count"]].groupby("CMPLNT_MONTH").sum().reset_index()
grouped_by_month = grouped_by_month[grouped_by_month["CMPLNT_MONTH"]>6]
grouped_by_month["group"] = grouped_by_month["CMPLNT_MONTH"].map(month_dict)
grouped_by_month["value"] = grouped_by_month["CMPLNT_NUM_count"]
return grouped_by_month[["group","value"]].to_json(orient="records")
@app.route("/tabledata/<indexids>",methods = ['GET'])
def get_table_data(indexids):
columns = ["neighbourhood_group","name","description","listing_url"]
json_indexids = json.dumps(indexids)
if json_indexids == "\"all\"":
top_3_default = stratified_sample.sort_values(['price', 'review_scores_rating','CMPLNT_NUM_count_sum'], ascending=[True, False, True])[columns].head(3)
return top_3_default.to_json(orient="records")
else:
stratified_sample_new = stratified_sample.set_index('index')
json_indexes = json.loads(json_indexids)[1:-1]
json_indexes = list(map(int,json_indexes.split(',')))
brushed_sample = stratified_sample_new[stratified_sample_new.index.isin(json_indexes)].sort_values(['price', 'review_scores_rating','CMPLNT_NUM_count_sum'], ascending=[True, False, True])[columns].head(3)
return brushed_sample.to_json(orient="records")
@app.route("/tabledatafilter/<boroughs>",methods = ['GET'])
def get_table_data_filter(boroughs):
columns = ["neighbourhood_group","name","description","listing_url"]
boroughs = boroughs.replace("\"","")
borough_list = json.dumps(boroughs)
borough_list = json.loads(borough_list)[1:-1]
borough_list = list(borough_list.split(','))
stratified_sample_new = stratified_sample.set_index('index')
brushed_sample = stratified_sample_new[stratified_sample_new["neighbourhood_group"].isin(borough_list)].sort_values(['price', 'review_scores_rating','CMPLNT_NUM_count_sum'], ascending=[True, False, True])[columns].head(3)
return brushed_sample.to_json(orient="records")