/
evaluate_Tracking_OneTrack.py
146 lines (106 loc) · 4.93 KB
/
evaluate_Tracking_OneTrack.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
import dgl
import dgl.function as fn
import torch.nn.functional as F
from torch.utils.data import Dataset, DataLoader, Sampler
import numpy as np
import h5py
import torch
import torch.nn as nn
import torch.optim as optim
import math
import uproot#3 as uproot
import numpy as np
import pandas as pd
from tqdm import tqdm
import csv
torch.manual_seed(0)
import os, sys
start_N = 359403
end_N = 359503
df_h = pd.read_csv('/data/wachan/ds100/event000001000-hits.csv')
df_d = pd.read_csv('/data/wachan/ds100/event000001000-doublets.csv')
os.environ["CUDA_VISIBLE_DEVICES"]="0"
cuda_device = torch.device('cuda:0' if torch.cuda.is_available() else 'cpu' )
print('cuda_device : ', cuda_device)
from modules.OneTrackDataloader import OneTrackDataset, collate_graphs
from modules.dynamic_graph import Dynamic_Graph_Model
path = '/data/wachan/ds100/OneTrackSamples/evt1000_Nearby_Hit_v4_Random/'
#50%
data_set_test = OneTrackDataset(path, num_start=359403, num_end=449260)
#100%
#data_set_test = OneTrackDataset(path, num_start=718803, num_end=898516)
#20%
#data_set_test = OneTrackDataset(path, num_start=143763, num_end=179706)
#Test
#data_set_test = OneTrackDataset(path, num_start=start_N, num_end=end_N)
test_loader = DataLoader(data_set_test, batch_size=128, shuffle=False,collate_fn=collate_graphs, num_workers=0)
model = Dynamic_Graph_Model(feature_dims_x = [4,32,16,3], feature_dims_en = [4,32,16,4])
model_name = 'model_DynamicGraphTrack_Score_NH_v4_50p_batch_size_128.pt'
out_name = 'PredictionFile_DynamicGraphTrack_Score_NH_v4_batch_size_128.h5'
model.to(cuda_device)
model.load_state_dict(torch.load(model_name))
param_numb = sum(p.numel() for p in model.parameters() if p.requires_grad)
print('Total parameters : ', param_numb)
model.eval()
truth_scores_Match, predicted_scores_Match, truth_scores_MisMatch, predicted_scores_MisMatch = [], [], [], []
T_QS_Match,T_QS_MisMatch,P_QS_Match,P_QS_MisMatch = [], [], [], []
Triplet_Sum = {}
with tqdm(test_loader, ascii=True) as tq:
count = start_N
for graph, label, score in tq:
print (count)
T_QS_Match_list,T_QS_MisMatch_list,P_QS_Match_list,P_QS_MisMatch_list = [], [], [], []
score = score.to(cuda_device)
pred_score = model(graph[0].to(cuda_device))[1]
#print (graph[0].ndata['x'][0][0])
for i in range(len(pred_score)):
#print (i)
truth_scores_Match.append(score[i][0].item())
truth_scores_MisMatch.append(score[i][1].item())
predicted_scores_Match.append(pred_score[i][0].item())
predicted_scores_MisMatch.append(pred_score[i][1].item())
T_QS_Match_list.append(score[i][0].item())
T_QS_MisMatch_list.append(score[i][1].item())
P_QS_Match_list.append(pred_score[i][0].item())
P_QS_MisMatch_list.append(pred_score[i][1].item())
#print (i)
#if (pred_score[i][0] > 0.85):
#print ()
#print ('Less than 0.9995, with pred score: ', pred_score[i][0].item(), ', target score: ', score[i][0].item(), ', ', score[i][1].item())
del graph; del score; del pred_score;
#print (T_QS_Match_list)
#print (sum(PScore_list)/2)
if len(T_QS_Match_list) == 6:
T_QS_Match.append(sum(T_QS_Match_list))
T_QS_MisMatch.append(sum(T_QS_MisMatch_list))
P_QS_Match.append(sum(P_QS_Match_list))
P_QS_MisMatch.append(sum(P_QS_MisMatch_list))
print (P_QS_Match)
hits= pd.read_csv(path + 'OneTrackSample00%s-hits.csv'%count)
doublets = pd.read_csv(path + 'OneTrackSample00%s-doublets.csv'%count)
#print (doublets)
print (doublets['start'][0])
#print (hits)
#Need to implement the formula to get the strength
Q_Strength = 0
Q_Pair_1 = str(doublets['start'][0])+'_'+str(doublets['start'][1])+'_'+str(doublets['start'][2])
Q_Pair_2 = str(doublets['end'][0])+'_'+str(doublets['end'][1])+'_'+str(doublets['end'][2])
#print (Q_Pair_1,Q_Pair_2,Q_Strength)
Triplet_Pair = "{},{}".format(Q_Pair_1,Q_Pair_2)
#print (Triplet_Pair)
Triplet_Sum[Triplet_Pair] = Q_Strength
#print (Triplet_Sum)
#if count == 359405:
#exit()
count += 1
#print (len(PScore_All))
hf = h5py.File(out_name, 'w')
hf.create_dataset('truth_scores_Match', data=np.array(truth_scores_Match))
hf.create_dataset('truth_scores_MisMatch', data=np.array(truth_scores_MisMatch))
hf.create_dataset('predicted_scores_Match', data=np.array(predicted_scores_Match))
hf.create_dataset('predicted_scores_MisMatch', data=np.array(predicted_scores_MisMatch))
hf.create_dataset('T_QS_Match', data=np.array(T_QS_Match))
hf.create_dataset('T_QS_MisMatch', data=np.array(T_QS_MisMatch))
hf.create_dataset('P_QS_Match', data=np.array(P_QS_Match))
hf.create_dataset('P_QS_MisMatch', data=np.array(P_QS_MisMatch))
hf.close()