forked from shermanash/DFSharp
-
Notifications
You must be signed in to change notification settings - Fork 0
/
optimizer_openopt.py
208 lines (186 loc) · 7.88 KB
/
optimizer_openopt.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
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
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
#!/usr/bin/python
'''
Simplest OpenOpt KSP example;
requires FuncDesigner installed.
For some solvers limitations on time, cputime, "enough" value, basic GUI features are available.
See http://openopt.org/KSP for more details
'''
import pandas as pd
from openopt import *
import math
import csv
from pprint import pprint
import datetime
from proj_elastic import InsertProj
from fuzzywuzzy import process
import numpy as np
def load_projections(projections_file):
projections = {}
with open(projections_file, 'r') as csvfile:
reader = csv.reader(csvfile)
for row in reader:
projections[row[0]] = float(row[1])
return projections
''' fuzzy match
inputs - list of strings typed by user
- list of options for players
outputs - list of fuzzy matched strings
'''
def fuzzy_match(inputlist, choices):
if (isinstance(inputlist, str)):
fz = process.extractOne(inputlist, choices)
return(fz[0])
else:
outlist = []
for s in inputlist:
fz = process.extractOne(s, choices)
outlist.append(fz[0])
return(outlist)
def optimize_grid(delta=[-4, -2, 0, 2, 4], min_dvp=[0, 1, 2, 3], min_sal=[2000, 2500,
3000], exclusions=[], locks=[], target=['DK_Proj', 'min_proj', 'proj_pure']):
df = pd.DataFrame()
for d in delta:
for v in min_dvp:
for m in min_sal:
for t in target:
players = optimizer(
delta=d,
min_dvp=v,
min_sal=m,
exclusions=exclusions,
locks=locks,
target=t,
return_format='df')
df = df.append(players)
return(df)
def optimizer(locks=[], exclusions=[], delta=0, min_own=0, min_dvp=0,
min_sal=2000, max_own=100, target='DK_Proj', return_format='ajax'):
adjustments = {}
items = []
player_ids = {}
today = datetime.datetime.today() - datetime.timedelta(hours=4)
yesterday = today - datetime.timedelta(hours=24)
path = '/home/ubuntu/dfsharp/opt_csvs/' + \
today.strftime('%Y%m%d') + '_opt.csv'
ypath = '/home/ubuntu/dfsharp/opt_csvs/' + \
yesterday.strftime('%Y%m%d') + '_opt.csv'
abblist = ['bkn', 'bos', 'cha', 'chi', 'cle', 'dal', 'den', 'det', 'hou',
'ind', 'lac', 'lal', 'mem', 'mia', 'mil', 'min', 'nor', 'nyk',
'okc', 'orl', 'phi', 'pho', 'por', 'sac', 'sas', 'tor', 'uta',
'was', 'atl', 'gsw']
# read csv of players
try:
df = pd.read_csv(path)
except IOError:
df = pd.read_csv(ypath)
path = ypath
playernames = df['name'].tolist()
teamnames = list(df['Team'].unique())
if len(locks) > 0:
locks = fuzzy_match(locks, playernames)
# loop through exclusions, fuzzy teams one way and players another
if len(exclusions) > 0:
combo = teamnames + playernames
exclusions = fuzzy_match(exclusions, combo)
with open(path, 'r') as csvfile:
reader = csv.reader(csvfile)
index = 0
for row in reader:
if (row[1] in locks) or ((index != 0) and (row[1] not in exclusions) and (row[11] not in exclusions) and (float(row[13]) >= float(min_own)) and (
float(row[13]) <= float(max_own)) and (float(row[16]) >= float(min_dvp)) and (float(row[4]) > 5) and (float(row[2]) > float(min_sal))):
vals = {
'id': index - 1,
'PG': 1 if row[0] == 'PG' else 0,
'SG': 1 if row[0] == 'SG' else 0,
'SF': 1 if row[0] == 'SF' else 0,
'PF': 1 if row[0] == 'PF' else 0,
'C': 1 if row[0] == 'C' else 0,
'name': row[1],
'salary': int(row[2]),
'DK_Proj': float(row[4]) if row[1] not in adjustments.keys() else adjustments[row[1]],
'min_proj': float(row[8]) if row[1] not in adjustments.keys() else adjustments[row[1]],
'dk_per_min': float(row[9]) if row[1] not in adjustments.keys() else adjustments[row[1]],
'lock': 1 if row[1] in locks else 0,
#'ownership': float(row[13]) if row[1] not in adjustments.keys() else adjustments[row[1]],
'dvprank': float(row[16]) if row[1] not in adjustments.keys() else adjustments[row[1]],
#'otprank': float(row[17]) if row[1] not in adjustments.keys() else adjustments[row[1]]
'usage_5g_avg': float(row[19]) if row[1] not in adjustments.keys() else adjustments[row[1]],
'value_3g_avg': float(row[28]) if row[1] not in adjustments.keys() else adjustments[row[1]],
'proj_pure': float(row[31]) if row[1] not in adjustments.keys() else adjustments[row[1]],
'dk_std_90_days': float(row[33]) if row[1] not in adjustments.keys() else adjustments[row[1]],
}
for team in abblist:
vals[team] = 1 if row[11] == team else 0
vals['PGSGC'] = vals['PG'] + vals['SG'] + vals['C']
vals['PFSFC'] = vals['PF'] + vals['SF'] + vals['C']
# if projections != None:
# vals['fpts'] = projections[vals['name']]
items.append(vals)
index += 1
for item in items:
for i in range(len(items)):
item['id%d' % i] = float(item['id'] == i)
constraints = lambda values: (
values['lock'] == len(locks),
values['salary'] >= 49500,
#values['salary'] != 50100,
values['salary'] <= int(50000),
values['nItems'] == 8,
values['PG'] >= 1,
values['PG'] <= 2,
values['SG'] >= 1,
values['SG'] <= 2,
values['SF'] >= 1,
values['SF'] <= 2,
values['PF'] >= 1,
values['PF'] <= 2,
values['C'] >= 1,
values['PFSFC'] >= 4,
values['PFSFC'] <= 5,
values['PGSGC'] >= 4,
values['PGSGC'] <= 5,
) + tuple([values['id%d' % i] <= 1 for i in range(len(items))])
#objective = lambda val: val['fpts'] + delta*val['otprank']
objective = lambda val: val[target] + delta * val['dk_std_90_days']
p = KSP(objective, items, goal='max', constraints=constraints)
# requires cvxopt and glpk installed, see http://openopt.org/KSP for other
# solvers
r = p.solve('glpk', iprint=0)
''' Results for Intel Atom 1.6 GHz:
------------------------- OpenOpt 0.50 -------------------------
solver: glpk problem: unnamed type: MILP goal: max
iter objFunVal log10(maxResidual)
0 0.000e+00 0.70
1 2.739e+01 -100.00
istop: 1000 (optimal)
Solver: Time Elapsed = 0.82 CPU Time Elapsed = 0.82
objFunValue: 27.389749 (feasible, MaxResidual = 0)
'''
# print(r.xf)
playerlist = r.xf
# r.xf is a list of players- we will merge their info back and return a DF
# instead
df2 = df[df['name'].isin(playerlist)]
# df2 is the latest lineup - we'll return the frame [for now]
df2[['DK_Proj', 'min_proj', 'dk_per_min', 'value', 'usage_5g_avg']] = np.round(
df2[['DK_Proj', 'min_proj', 'dk_per_min', 'value', 'usage_5g_avg']], 1)
ajax = df2[['numpos',
'name',
'Team',
'Opp',
'dk_sal',
'ownership',
'DK_Proj',
'dvprank',
'min_proj',
'dk_per_min',
'value',
'usage_5g_avg']].to_json(orient='records')
#ajax = df2.to_html(index=False)
InsertProj(df2, indexer="latestlineup")
# return(playerlist)
print(constraints)
if(return_format == 'ajax'):
return(ajax)
else:
return(df2)