GRASPoptimization/plot_po.py at master · devincody/GRASPoptimization · GitHub

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
import pandas as pd
import numpy as np

import matplotlib.pyplot as plt
import os

start_f = 0.38261
end_f = 2.6087
n_f = 66

report_top = 10

po_path = "../po_opt"
if not os.path.exists(po_path):
	os.mkdir(po_path)

os.chdir(po_path)


cwd = os.getcwd()
items = os.listdir(cwd)
for it in items:
	if ".csv" in it[-5:]:
		print("Processing file ", it)
		if "QRFH" in it:
			start_f = 0.38261
			end_f = 2.6087
			n_f = 66
		elif "DSA" in it:
			start_f = 1.05
			end_f = 2.0
			n_f = 20

		plt.figure()
		ax = plt.axes()

		# Make internal folder for results
		path = it[:-4]
		if not os.path.exists(path):
			os.mkdir(path)

		# Read csv
		a = pd.read_csv(it, skiprows=2, index_col=False)
		a.dropna(inplace = True) #get rid of nans

		a["Efficiency"] *= -100

		# Find which values of z_dist are used
		z_dists = a["z_dist"].unique()
		number_of_z_dists = len(z_dists)

		# Append frequency information to dataFrame
		freq = np.linspace(start_f,end_f,n_f)
		for i in range(len(a)):
			a.at[i, "f"] = freq[int(i/number_of_z_dists)]


		labels = [] #keep track of z_dists which make the cutoff
		avg_efficiencies = [] # list for determining cutoff
		good_dists = pd.DataFrame(columns = a.keys()) # DataFrame for top $(report_top) results

		# Find best curves
		for i in z_dists:
			one_z_dist = a.where(a["z_dist"]==i)
			one_z_dist.dropna(inplace = True)

			#append tuples with (z_dist, avg(eff))
			avg_efficiencies.append((i,np.mean(one_z_dist["Efficiency"])))

		# Sort Efficiencies and take the $(report_top)th best efficiency
		avg_efficiencies.sort(key=lambda x:x[1], reverse=True)
		best_configs = [x[0] for x in avg_efficiencies[:report_top]]
		# best_configs.reverse() #start at smallest

		for i in best_configs[::-1]: # iterate backwards (most efficient last,
									 # so it appears first in .csv)

			# For each z_dist, create smaller dataFrame with only the one z_dist
			one_z_dist = a.where(a["z_dist"]==i)
			one_z_dist.dropna(inplace = True)

			one_z_dist.plot(x="f", y="Efficiency", ax = ax)
			labels.append("%6.3f"%i)
			good_dists = one_z_dist.append(good_dists)

		# Format and Save
		ax.legend(labels)
		ax.set_title("Efficiency vs. Frequency at several Focus Lengths")
		ax.set_xlabel("Frequency [GHz]")
		ax.set_ylabel("Efficiency [%]")
		plt.ylim(0,80)
		plt.savefig(path + "/" + path + ".png")

		# Remove "Unnamed" column in pandas Dataframe
		keys = good_dists.keys()
		try:
			if "Unnamed" in keys[-2]:
				good_dists = good_dists.drop(keys[-2], axis = 1)
				print("dropping unnamed column", keys[-2])
		except:
			pass

		# good_dists.sort_values(by="f", inplace=True)
		good_dists.to_csv("%s/%s_best_focal_lengths.csv"%(path,path))