-
Notifications
You must be signed in to change notification settings - Fork 2
/
settings.toml
119 lines (105 loc) · 3.17 KB
/
settings.toml
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
[default]
# Required config to connect to the MongoDB Database
username = ""
password = ""
port = 27017
host = "localhost"
master_database = "master"
database = "db"
query_database = "query_db"
# Path to the dataset
data_path = "../Data/phase3_sample_data/Labelled/Set2"
# Path to unlabled dataset
unlabeled_data_path = "../Data/phase3_sample_data/Unlabelled/Set 2"
# Path to all 11k hands (Task 5 and 6)
master_data_path = "../Data/Hands/"
# Path to the output
output_path = "../Outputs/"
# Path to templates
template_path = "templates"
# Path to metadata
metadata_csv = "../Data/phase3_sample_data/labelled_set2.csv"
# metadata_csv = "../Data/HandInfo.csv"
# Path to unlabeled metadata
unlabeled_metadata_csv = "../Data/phase3_sample_data/unlabelled_set2.csv"
# Path to 11k metadata csv
master_metadata_csv = "../Data/HandInfo.csv"
task_five_output = "task_five_output"
[default.ppr]
[default.ppr.task_3]
cache_dir = '../Cache/PPR/Task3'
[default.ppr.feedback]
cache_dir = "../Cache/PPR/Feedback"
[default.ppr.classifier]
frt = 'nmf'
# k=58 works great for math_method
k = 25
feature = 'moment_inv'
edges = 90
alpha = 0.15
convergence_method = 'power_iteration'
# ignoring metadata gives good results for set2 data and power iteration
ignore_metadata = false
[default.decision.classifier]
model = 'moment_inv'
max_depth = 15
min_size = 30
[default.svm.classifier]
k = 10
frt = 'nmf'
model = 'moment'
# The feature models
[default.sift]
# The colletion to store sift data
collection = "img_sift"
# Use opencvs implementation of SIFT algorithm or use the binary provided by
# David Lowe
use_opencv = 1
bin_path = "./bin/sift"
kmeans_verbosity = 0
# [default.moment]
# # The colletion to store moments data
# collection = "img_moment"
# collection_inv = "img_moment_inv"
# # Weights used for the calculation of Moment
# # Format is - W_Y_1, W_Y_2, W_Y_3, W_U_1, W_U_2, W_U_3, W_V_1, W_V_2, W_V_3
# # Must be in double format.
# weights = [3.0, 3.0, 3.0, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5]
# Application configuration
[default.loader]
# Batch to operate on
batch_size = 1000
[default.window]
# Images are split into windows when calculating Image Moments. The two
# configuration specifies the dimensions of the window.
win_height = 100
win_width = 100
[default.moment]
# The colletion to store moments data
collection = "img_moment"
collection_inv = "img_moment_inv"
# Weights used for the calculation of Moment
# Format is - W_Y_1, W_Y_2, W_Y_3, W_U_1, W_U_2, W_U_3, W_V_1, W_V_2, W_V_3
# Must be in double format.
weights = [3.0, 3.0, 3.0, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5]
[default.hog]
collection = "img_hog"
scale_factor = 10
[default.images]
metadata_collection = 'metadata'
[default.lbp]
collection = 'lbp'
[default.task1_config]
model = 'sift'
frt = 'pca'
[default.task2_config]
model = 'moment'
max_iter = 500
[default.task5_config]
max_evals = 1000
gamma_max = 0
lda_max_evals = 5
lda_gamma_max = 0
nmf_gamma_max = 2000
nmf_max_evals = 500
clfs_directory = '../Outputs/clfs/'