Newer
Older
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
import numpy as np
import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn.pipeline import make_pipeline
from sklearn.ensemble import RandomForestRegressor
from sklearn.preprocessing import StandardScaler
from sklearn.metrics import mean_absolute_error, mean_squared_error
def load_split(filepath):
soc = pd.read_csv(filepath, header=0)
X_soc = soc[["Catch", "Conv", "Elev", "NDVI_median"]] # MAKE THIS AS AN INPUT TO THE FUNCTION TO CHOOSE WHICH PARAMS WE TRAIN WITH
y_soc = soc["SOC (%)"]
X_train, X_test, y_train, y_test = train_test_split(X_soc, y_soc, test_size=0.2, random_state=1)
X_train, X_val, y_train, y_val = train_test_split(X_train, y_train, test_size = 0.25, random_state=1)
return X_train, X_val, X_test, y_train, y_val, y_test
def RFreg_test(X_train, X_test, y_train, y_test):
pipeline = RandomForestRegressor(n_estimators=100, random_state=0)
pipeline.fit(X_train, y_train)
y_pred = pipeline.predict(X_test)
print("Mean Squared Error is: %.2f"%(mean_squared_error(y_test, y_pred)))
print("Mean Absolute Error is: %.2f"%(mean_absolute_error(y_test, y_pred)))
print("")
return 0
def main():
# Initial test
"""PART 1 - LOADING IN THE DATA"""
# Load in the training, validation and test from dataset
X_tr, X_va, X_te, y_tr, y_va, y_te = load_split('Project/SOC_Database.csv')
"""PART 2 - CHOOSING FEATURES TO TRAIN ON"""
# Make a section where we will be testing with various different parameters in the training set (variables)
# To obtain the optimal variables to train with
"""PART 3 - VALIDATION TESTING TO FIND THE OPTIMAL PARAMETERS FOR EACH ALGORITHM"""
# Make a section where we will be testing with the validation set and choosing the best parameters
# WRITE CODE
# STEPS FOR VALIDATION TESTING TO FIND OPTIMAL PARAMETERS:
# i) Choose 5 algorithms that we want to test (we can also do more!)
# ii) Identify the parameters for each algorithm that we need to optimize
# iii) Choose an evaluation method (rmse, mae, determination coefficient, F1-score, ratio of performance to deviation)
# iv) Estimate a logarithmic range of 10 values for each parameter and run loop testing combinations
# of those parameters with eachother and downselect the combo of parameters with smallest error
# v) Select for each algorithm the optimal parameters for the next section
"""PART 4 - RUN TEST SET FOR EACH ALGORITHM WITH THE OPTIMAL PARAMETERS"""
# Test algorithms again with the best parameters chosen from previous section to do a direct comparison
# Test each algorithm
RFreg_test(X_tr, X_te, y_tr, y_te)
"""PART 5 - TBD: IF ENOUGH TIME AT HAND: REPEAT TESTING IN A DIFFERENT DATASET AND COMPARE PERFORMANCE"""
# This a test
# new
if __name__ == "__main__":
main()