# Eigen-Adaptive
'F_min' : 0.9,
'epsilon' : 0.05,
# Pedro Adaptive
'e_low' : 0.95,
'e_high' : 0.98,
'static_r' : 0,
'r_upper_bound' : None,
'fix_init_Q' : 0,
'small_value' : 0.0001,
'ignoreUp2' : 0 }
p['x_thresh'] = sp.stats.t.isf(0.5* p['FP_rate'], p['sample_N'])
''' Load Data '''
data = load_ts_data('isp_routers', 'full')
#data, sins = sin_rand_combo(5, 1000, [10, 35, 60], noise_scale = 0.2, seed = 1)
data = zscore(data)
z_iter = iter(data)
numStreams = data.shape[1]
'''Initialise'''
st = initialise(p, numStreams)
'''Begin Frahst'''
# Main iterative loop.
for zt in z_iter:
zt = zt.reshape(zt.shape[0],1) # Convert to a column Vector
st['anomaly'] = False
'''Frahst Version '''
# Eigen-Adaptive
'F_min': 0.9,
'epsilon': 0.05,
# Pedro Adaptive
'e_low': 0.95,
'e_high': 0.98,
'static_r': 0,
'r_upper_bound': None,
'fix_init_Q': 0,
'small_value': 0.0001,
'ignoreUp2': 0
}
p['x_thresh'] = sp.stats.t.isf(0.5 * p['FP_rate'], p['sample_N'])
''' Load Data '''
data = load_ts_data('isp_routers', 'full')
#data, sins = sin_rand_combo(5, 1000, [10, 35, 60], noise_scale = 0.2, seed = 1)
data = zscore(data)
z_iter = iter(data)
numStreams = data.shape[1]
'''Initialise'''
st = initialise(p, numStreams)
'''Begin Frahst'''
# Main iterative loop.
for zt in z_iter:
zt = zt.reshape(zt.shape[0], 1) # Convert to a column Vector
st['anomaly'] = False
'''Frahst Version '''
st = FRAHST_V7_0_iter(zt, st, p)
# Calculate reconstructed data if needed
"M": 5,
"pA": 0.1,
"noise_sig": 0.1,
"seed": None,
}
anomaly_type = "peak_dip"
gen_funcs = dict(
peak_dip=gen_a_peak_dip, grad_persist=gen_a_grad_persist, step=gen_a_step, trend=gen_a_periodic_shift
)
# D = gen_funcs[anomaly_type](**a)
# data = D['data']
data = load_ts_data("isp_routers", "full")
# execfile('/Users/chris/Dropbox/Work/MacSpyder/Utils/gen_simple_peakORshift_data.py')
# data = B
""" Mean Centering """
# data = zscore_win(data, 100)
data = zscore(data)
z_iter = iter(data)
numStreams = data.shape[1]
"""Initialise"""
Frahst_alg = FRAHST("F-7.A-recS.R-static", p, numStreams)
"""Begin Frahst"""
# Main iterative loop.
for zt in z_iter:
step = gen_a_step) """ Choice of data sets. Comment/Uncomment to choose. """ """----------------------------------------------------""" ''' Synthetic Data sets ''' #data_name = 'synth' #D = gen_funcs[anomaly_type](**a) #raw_data = D['data'] #data = raw_data.copy() '''ISP data sets ''' data_name = 'isp_routers' raw_data = load_ts_data(data_name, 'full') data = raw_data.copy() ''' Sensor Motes data sets ''' #data_name = 'motes_l' #raw_data = load_data(data_name) #data = clean_zeros(raw_data, cpy=1) ''' Data Preprocessing ''' """ Data is loaded into memory, mean centered and standardised then converted to an iterable to read by the CD-ST each iteration""" #data = zscore_win(data, 100) # Sliding window implimentation data = zscore(data) # Batch method implimentation