def runner(config):
generate_args, analyze_args, inf_seed = _munge_config(config)
# generate synthetic data
T, M_c, M_r, X_L, X_D = du.generate_clean_state(max_mean=10, max_std=1,
**generate_args)
table_shape = map(len, (T, T[0]))
start_dims = du.get_state_shape(X_L)
# run engine with do_timing = True
engine = LocalEngine(inf_seed)
X_L, X_D, (elapsed_secs,) = engine.analyze(M_c, T, X_L, X_D,
do_timing=True,
**analyze_args
)
#
end_dims = du.get_state_shape(X_L)
same_shape = start_dims == end_dims
summary = dict(
elapsed_secs=elapsed_secs,
same_shape=same_shape,
)
ret_dict = dict(
config=config,
summary=summary,
table_shape=table_shape,
start_dims=start_dims,
end_dims=end_dims,
)
return ret_dict
def gen_data(**kwargs):
T, M_c, M_r, gen_X_L, gen_X_D = du.generate_clean_state(**kwargs)
#
engine = LocalEngine()
sampled_T = gu.sample_T(engine, M_c, T, gen_X_L, gen_X_D)
T_test = random.sample(sampled_T, n_test)
gen_data_ll = ctu.calc_mean_test_log_likelihood(M_c, T, gen_X_L, gen_X_D, T)
gen_test_set_ll = ctu.calc_mean_test_log_likelihood(M_c, T, gen_X_L, gen_X_D, T_test)
#
return T, M_c, M_r, T_test, gen_data_ll, gen_test_set_ll
def gen_data(**kwargs):
T, M_c, M_r, gen_X_L, gen_X_D = du.generate_clean_state(**kwargs)
#
engine = LocalEngine()
sampled_T = gu.sample_T(engine, M_c, T, gen_X_L, gen_X_D)
T_test = random.sample(sampled_T, n_test)
gen_data_ll = ctu.calc_mean_test_log_likelihood(M_c, T, gen_X_L, gen_X_D, T)
gen_test_set_ll = ctu.calc_mean_test_log_likelihood(M_c, T, gen_X_L, gen_X_D, T_test)
#
return T, M_c, M_r, T_test, gen_data_ll, gen_test_set_ll
def draw_a_cc_state(filename):
rng_seed = random.randrange(10000)
num_rows = 100
num_cols = 50
num_splits = 5
num_clusters = 5
nan_prop = .25
table_name = 'plottest'
generator_name = 'plottest_cc'
# generate some clustered data
ccmd = du.generate_clean_state(rng_seed, num_clusters, num_cols, num_rows,
num_splits)
T, _M_c, _M_r, _X_L, _X_D = ccmd
for row in range(num_rows):
for col in range(num_cols):
if random.random() < nan_prop:
T[row][col] = float('nan')
input_df = pd.DataFrame(T, columns=['col_%i' % i for i in range(num_cols)])
os.environ['BAYESDB_WIZARD_MODE'] = '1'
bdb = bayeslite.bayesdb_open()
bayesdb_read_pandas_df(bdb, table_name, input_df, create=True)
bdb.execute('''
create generator {} for {} using crosscat(guess(*))
'''.format(generator_name, table_name))
bdb.execute('initialize 4 models for {}'.format(generator_name))
bdb.execute('analyze {} for 10 iterations wait'.format(generator_name))
plt.figure(facecolor='white', tight_layout=False)
draw_state(
bdb,
'plottest',
'plottest_cc',
0,
separator_width=1,
separator_color=(0., 0., 1., 1.),
short_names=False,
nan_color=(1, .15, .25, 1.))
plt.savefig(filename)
def draw_a_cc_state(filename):
rng_seed = random.randrange(10000)
num_rows = 100
num_cols = 50
num_splits = 5
num_clusters = 5
nan_prop = .25
table_name = 'plottest'
generator_name = 'plottest_cc'
# generate some clustered data
ccmd = du.generate_clean_state(rng_seed, num_clusters, num_cols, num_rows,
num_splits)
T, _M_c, _M_r, _X_L, _X_D = ccmd
for row in range(num_rows):
for col in range(num_cols):
if random.random() < nan_prop:
T[row][col] = float('nan')
input_df = pd.DataFrame(T, columns=['col_%i' % i for i in range(num_cols)])
os.environ['BAYESDB_WIZARD_MODE']='1'
bdb = bayeslite.bayesdb_open()
bayesdb_read_pandas_df(bdb, table_name, input_df, create=True)
bdb.execute('''
create generator {} for {} using crosscat(guess(*))
'''.format(generator_name, table_name))
bdb.execute('initialize 4 models for {}'.format(generator_name))
bdb.execute('analyze {} for 10 iterations wait'.format(generator_name))
plt.figure(facecolor='white', tight_layout=False)
draw_state(bdb, 'plottest', 'plottest_cc', 0,
separator_width=1, separator_color=(0., 0., 1., 1.),
short_names=False, nan_color=(1, .15, .25, 1.))
plt.savefig(filename)