tangram_linear 0.7.0

Tangram makes it easy for programmers to train, deploy, and monitor machine learning models.
Documentation 
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
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165

from pandas.api.types import CategoricalDtype
from sklearn.metrics import accuracy_score, roc_auc_score
from sklearn.metrics import mean_squared_error
import argparse
import numpy as np
import pandas as pd
import json

parser = argparse.ArgumentParser()
parser.add_argument('--library', choices=['sklearn', 'pytorch'], required=True)
args = parser.parse_args()

# Load the data.
# path_train = 'data/flights_100k_train.csv'
# path_test = 'data/flights_test.csv'
# path_train = 'data/flights_1m_train.csv'
# path_test = 'data/flights_test.csv'
path_train = 'data/flights_10m_train.csv'
path_test = 'data/flights_test.csv'
target_column_name = "dep_delayed_15min"
month_options = [
	"c-1", "c-10", "c-11", "c-12", "c-2", "c-3", "c-4", "c-5", "c-6", "c-7", "c-8", "c-9",
]
day_of_week_options = ["c-1", "c-2", "c-3", "c-4", "c-5", "c-6", "c-7"]
day_of_month_options = [
	"c-1", "c-10", "c-11", "c-12", "c-13", "c-14", "c-15", "c-16", "c-17", "c-18", "c-19",
	"c-2", "c-20", "c-21", "c-22", "c-23", "c-24", "c-25", "c-26", "c-27", "c-28", "c-29",
	"c-3", "c-30", "c-31", "c-4", "c-5", "c-6", "c-7", "c-8", "c-9",
]
carrier_options = [
	"AA", "AQ", "AS", "B6", "CO", "DH", "DL", "EV", "F9", "FL", "HA", "HP", "MQ", "NW", "OH",
	"OO", "TZ", "UA", "US", "WN", "XE", "YV",
]
origin_options = [
	"ABE", "ABI", "ABQ", "ABY", "ACK", "ACT", "ACV", "ACY", "ADK", "ADQ", "AEX", "AGS", "AKN",
	"ALB", "AMA", "ANC", "APF", "ASE", "ATL", "ATW", "AUS", "AVL", "AVP", "AZO", "BDL", "BET",
	"BFL", "BGM", "BGR", "BHM", "BIL", "BIS", "BLI", "BMI", "BNA", "BOI", "BOS", "BPT", "BQK",
	"BQN", "BRO", "BRW", "BTM", "BTR", "BTV", "BUF", "BUR", "BWI", "BZN", "CAE", "CAK", "CDC",
	"CDV", "CEC", "CHA", "CHO", "CHS", "CIC", "CID", "CLD", "CLE", "CLL", "CLT", "CMH", "CMI",
	"COD", "COS", "CPR", "CRP", "CRW", "CSG", "CVG", "CWA", "DAB", "DAL", "DAY", "DBQ", "DCA",
	"DEN", "DFW", "DHN", "DLG", "DLH", "DRO", "DSM", "DTW", "EGE", "EKO", "ELP", "ERI", "EUG",
	"EVV", "EWR", "EYW", "FAI", "FAR", "FAT", "FAY", "FCA", "FLG", "FLL", "FLO", "FNT", "FSD",
	"FSM", "FWA", "GEG", "GFK", "GGG", "GJT", "GNV", "GPT", "GRB", "GRK", "GRR", "GSO", "GSP",
	"GST", "GTF", "GTR", "GUC", "HDN", "HKY", "HLN", "HNL", "HOU", "HPN", "HRL", "HSV", "HTS",
	"HVN", "IAD", "IAH", "ICT", "IDA", "ILG", "ILM", "IND", "IPL", "ISO", "ISP", "ITO", "IYK",
	"JAC", "JAN", "JAX", "JFK", "JNU", "KOA", "KTN", "LAN", "LAS", "LAW", "LAX", "LBB", "LCH",
	"LEX", "LFT", "LGA", "LGB", "LIH", "LIT", "LNK", "LRD", "LSE", "LWB", "LWS", "LYH", "MAF",
	"MBS", "MCI", "MCN", "MCO", "MDT", "MDW", "MEI", "MEM", "MFE", "MFR", "MGM", "MHT", "MIA",
	"MKE", "MLB", "MLI", "MLU", "MOB", "MOD", "MOT", "MQT", "MRY", "MSN", "MSO", "MSP", "MSY",
	"MTJ", "MYR", "OAJ", "OAK", "OGG", "OKC", "OMA", "OME", "ONT", "ORD", "ORF", "OTZ", "OXR",
	"PBI", "PDX", "PFN", "PHF", "PHL", "PHX", "PIA", "PIE", "PIH", "PIT", "PNS", "PSC", "PSE",
	"PSG", "PSP", "PVD", "PWM", "RAP", "RDD", "RDM", "RDU", "RFD", "RIC", "RNO", "ROA", "ROC",
	"RST", "RSW", "SAN", "SAT", "SAV", "SBA", "SBN", "SBP", "SCC", "SCE", "SDF", "SEA", "SFO",
	"SGF", "SGU", "SHV", "SIT", "SJC", "SJT", "SJU", "SLC", "SMF", "SMX", "SNA", "SOP", "SPI",
	"SPS", "SRQ", "STL", "STT", "STX", "SUN", "SWF", "SYR", "TEX", "TLH", "TOL", "TPA", "TRI",
	"TTN", "TUL", "TUP", "TUS", "TVC", "TWF", "TXK", "TYR", "TYS", "VCT", "VIS", "VLD", "VPS",
	"WRG", "WYS", "XNA", "YAK", "YUM",
]
dest_options= [
	"ABE", "ABI", "ABQ", "ABY", "ACK", "ACT", "ACV", "ACY", "ADK", "ADQ", "AEX", "AGS", "AKN",
	"ALB", "AMA", "ANC", "APF", "ASE", "ATL", "ATW", "AUS", "AVL", "AVP", "AZO", "BDL", "BET",
	"BFL", "BGM", "BGR", "BHM", "BIL", "BIS", "BLI", "BMI", "BNA", "BOI", "BOS", "BPT", "BQK",
	"BQN", "BRO", "BRW", "BTM", "BTR", "BTV", "BUF", "BUR", "BWI", "BZN", "CAE", "CAK", "CDC",
	"CDV", "CEC", "CHA", "CHO", "CHS", "CIC", "CID", "CLD", "CLE", "CLL", "CLT", "CMH", "CMI",
	"COD", "COS", "CPR", "CRP", "CRW", "CSG", "CVG", "CWA", "DAB", "DAL", "DAY", "DBQ", "DCA",
	"DEN", "DFW", "DHN", "DLG", "DLH", "DRO", "DSM", "DTW", "EGE", "EKO", "ELP", "ERI", "EUG",
	"EVV", "EWR", "EYW", "FAI", "FAR", "FAT", "FAY", "FCA", "FLG", "FLL", "FLO", "FNT", "FSD",
	"FSM", "FWA", "GEG", "GFK", "GGG", "GJT", "GNV", "GPT", "GRB", "GRK", "GRR", "GSO", "GSP",
	"GST", "GTF", "GTR", "GUC", "HDN", "HKY", "HLN", "HNL", "HOU", "HPN", "HRL", "HSV", "HTS",
	"HVN", "IAD", "IAH", "ICT", "IDA", "ILG", "ILM", "IND", "IPL", "ISO", "ISP", "ITO", "IYK",
	"JAC", "JAN", "JAX", "JFK", "JNU", "KOA", "KTN", "LAN", "LAS", "LAW", "LAX", "LBB", "LBF",
	"LCH", "LEX", "LFT", "LGA", "LGB", "LIH", "LIT", "LNK", "LRD", "LSE", "LWB", "LWS", "LYH",
	"MAF", "MBS", "MCI", "MCN", "MCO", "MDT", "MDW", "MEI", "MEM", "MFE", "MFR", "MGM", "MHT",
	"MIA", "MKE", "MLB", "MLI", "MLU", "MOB", "MOD", "MOT", "MQT", "MRY", "MSN", "MSO", "MSP",
	"MSY", "MTJ", "MYR", "OAJ", "OAK", "OGG", "OKC", "OMA", "OME", "ONT", "ORD", "ORF", "OTZ",
	"OXR", "PBI", "PDX", "PFN", "PHF", "PHL", "PHX", "PIA", "PIE", "PIH", "PIT", "PNS", "PSC",
	"PSE", "PSG", "PSP", "PVD", "PWM", "RAP", "RDD", "RDM", "RDU", "RFD", "RIC", "RNO", "ROA",
	"ROC", "RST", "RSW", "SAN", "SAT", "SAV", "SBA", "SBN", "SBP", "SCC", "SCE", "SDF", "SEA",
	"SFO", "SGF", "SGU", "SHV", "SIT", "SJC", "SJT", "SJU", "SLC", "SMF", "SMX", "SNA", "SOP",
	"SPI", "SPS", "SRQ", "STL", "STT", "STX", "SUN", "SWF", "SYR", "TEX", "TLH", "TOL", "TPA",
	"TRI", "TTN", "TUL", "TUP", "TUS", "TVC", "TWF", "TXK", "TYR", "TYS", "VCT", "VIS", "VLD",
	"VPS", "WRG", "WYS", "XNA", "YAK", "YUM",
]
dtype = {
	'month': CategoricalDtype(categories=month_options) ,
	'day_of_month': CategoricalDtype(categories=day_of_month_options),
	'day_of_week': CategoricalDtype(categories=day_of_week_options),
	'dep_time': np.int64,
	'unique_carrier': CategoricalDtype(categories=carrier_options),
	'origin': CategoricalDtype(categories=origin_options),
	'dest': CategoricalDtype(categories=origin_options),
	'distance': np.int64,
	'dep_delayed_15min': CategoricalDtype(categories=['N','Y']),
}
data_train = pd.read_csv(path_train, dtype=dtype)
data_test = pd.read_csv(path_test, dtype=dtype)
if args.library == 'pytorch' or args.library == 'sklearn':
	categorical_columns = data_train.select_dtypes(['category']).columns
	data_train.loc[:, categorical_columns] = data_train.loc[:, categorical_columns].apply(lambda x: x.cat.codes)
	data_test.loc[:, categorical_columns] = data_test.loc[:, categorical_columns].apply(lambda x: x.cat.codes)
features_train = data_train.loc[:, data_train.columns != target_column_name]
labels_train = data_train[target_column_name]
features_test = data_test.loc[:, data_test.columns != target_column_name]
labels_test = data_test[target_column_name]

if args.library == 'pytorch' or args.library == 'sklearn':
	from sklearn.linear_model import SGDRegressor
	from sklearn.preprocessing import StandardScaler
	from sklearn.compose import ColumnTransformer
	from sklearn.pipeline import Pipeline
	from sklearn.impute import SimpleImputer
	from sklearn.preprocessing import StandardScaler, OneHotEncoder
	numeric_features = features_train.select_dtypes(
		include=[np.float64, np.int64]
	).columns
	numeric_transformer = Pipeline(steps=[
		('imputer', SimpleImputer(strategy='median')),
		('scaler', StandardScaler())
	])
	categorical_features = features_train.select_dtypes(
		include=['category']
	).columns
	categorical_transformer = Pipeline(
		steps=[
			('imputer', SimpleImputer(strategy='constant', fill_value='missing')),
			('onehot', OneHotEncoder(handle_unknown='ignore'))
	])
	preprocessor = ColumnTransformer(
		transformers=[
			('num', numeric_transformer, numeric_features),
			('cat', categorical_transformer, categorical_features)
	])
	features_train = preprocessor.fit_transform(features_train)
	features_test = preprocessor.transform(features_test)

# Train the model.
if args.library == 'pytorch':
	from pytorch_linear import LinearBinaryClassifier
	model = LinearBinaryClassifier(batch_size=1000, n_epochs=1, learning_rate=0.01)
	model.fit(features_train, labels_train)
elif args.library == 'sklearn':
	from sklearn.linear_model import SGDClassifier
	model = SGDClassifier(
		max_iter=1,
		eta0=0.01,
		learning_rate='constant',
		tol=None,
		loss='log'
	)
	model.fit(features_train, labels_train)
elif args.library == 'tensorflow':
	pass

# Make predictions on the test data.
if args.library == 'pytorch':
	predictions_proba = model.predict_proba(features_test)
else:
	predictions_proba = model.predict_proba(features_test)[:, 1]

# Compute metrics.
auc_roc = roc_auc_score(labels_test, predictions_proba)

print(json.dumps({
	'auc_roc': auc_roc,
}))