-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathsubmission.py
121 lines (98 loc) · 4.99 KB
/
submission.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
import numpy as np
import pandas as pd
from sklearn.svm import SVR
from sklearn.metrics import mean_absolute_error
def generate_new_df_by_dict(old_df):
columns = old_df.columns.values[1:] # 旧列标
new_df_columns = ['day'] # 生成新df的列index
count = 30
for name in columns:
curr_name_list = []
while count > 0:
curr_name_list.append(f'{name}-{count}')
count -= 1
count = 30
new_df_columns = new_df_columns + curr_name_list
new_df_dict = {column_name: [] for column_name in new_df_columns} # 生成511个key的字典
# len(old_df) + 1，还有第一行是index
for i in range(30, len(old_df)): # 前0-29提供信息，从第30行开始
pre_thirty = old_df.iloc[i - 30: i] # 取下old_df前30行
new_df_dict['day'].append(i + 1) # 先加上day
for index, curr_column in pre_thirty.iteritems(): # 按列遍历
if index == 'day':
continue
count = 30
the_column = list(curr_column)
for j in range(len(the_column)):
new_df_dict[f'{index}-{count}'].append(the_column[j])
count -= 1
new_df = pd.DataFrame(new_df_dict)
return new_df
# 根据要求的feature和interval切割
def cut_by_feature_and_interval(original_df, feature_name_list, weather_interval, cases_interval):
weather_start = weather_interval # 每个weather feature开始的位置
case_start = cases_interval # 每个daily_case开始的位置
columns = feature_name_list
weather_features = list(columns[:len(columns) - 1]) # weather_features 名字 list
dailly_cases = columns[len(columns) - 1] # weather_features 名字 string
new_df_columns = [] # 生成新df的列column index
weather_count = weather_start # weather name 的计数器
for name in weather_features: # 除了dailly cases的weather feature name list
curr_name_list = []
while weather_count > 0:
curr_name_list.append(f'{name}-{weather_count}')
weather_count -= 1
weather_count = weather_start # 重置计数器
new_df_columns = new_df_columns + curr_name_list
daily_case_count = case_start # daily_case的计数器
daily_case_name_list = []
while daily_case_count > 0:
daily_case_name_list.append(f'{dailly_cases}-{daily_case_count}')
daily_case_count -= 1
new_df_columns = new_df_columns + daily_case_name_list
new_df_dict = {column_name: [] for column_name in new_df_columns} # 生成511个key的字典
for index, curr_columns in original_df.iteritems(): # 按列遍历
the_column = list(curr_columns)
if index in new_df_dict.keys():
new_df_dict[index] = the_column
new_df = pd.DataFrame(new_df_dict)
return new_df
## Project-Part1
def predict_COVID_part1(svm_model, train_df, train_labels_df, past_cases_interval, past_weather_interval, test_feature):
new_train_df = generate_new_df_by_dict(train_df) # 生成大的train_df
train_X = cut_by_feature_and_interval(new_train_df, ["max_temp", "max_dew", "max_humid", "dailly_cases"],
past_weather_interval,
past_cases_interval)
test_X_dict = dict()
for index, item in test_feature.iteritems():
if index != "day":
test_X_dict[index] = [item]
test_X_before_cut = pd.DataFrame(test_X_dict) # 输入的是列，把它变成行
test_X = cut_by_feature_and_interval(test_X_before_cut, ["max_temp", "max_dew", "max_humid", "dailly_cases"],
past_weather_interval,
past_cases_interval)
train_Y = train_labels_df.loc[30:, 'dailly_cases'] # 30 -192 天的dailly_cases
model = svm_model.fit(train_X, train_Y)
test_Y = model.predict(test_X)
return int(test_Y)
## Project-Part2
def predict_COVID_part2(train_df, train_labels_df, test_feature):
new_train_df = generate_new_df_by_dict(train_df) # 生成大的train_df
new_train_df = new_train_df.iloc[73:]
train_X = cut_by_feature_and_interval(new_train_df, ['max_temp', 'avg_temp', 'dailly_cases'],
30, 15)
test_X_dict = dict()
for index, item in test_feature.iteritems():
if index != "day":
test_X_dict[index] = [item]
test_X_before_cut = pd.DataFrame(test_X_dict) # 输入的是列，把它变成行
test_X = cut_by_feature_and_interval(test_X_before_cut,
['max_temp', 'avg_temp', 'dailly_cases'],
30, 15)
train_Y = train_labels_df.loc[103:, 'dailly_cases'] # 73 -192 天的dailly_cases
svm_model = SVR()
svm_model.set_params(**{'kernel': "poly", 'degree': 1, 'C': 15000,
'gamma': 'scale', 'coef0': 0.0, 'tol': 0.001, 'epsilon': 1})
model = svm_model.fit(train_X, train_Y)
test_Y = model.predict(test_X)
return int(test_Y)