def run_pcmci(data, tau_max=8, plot_res=False): var_names = list(data.columns) pcmci = pcmci_setup(data) # correlations = pcmci.get_lagged_dependencies(tau_max=20, val_only=True)['val_matrix'] results = pcmci.run_pcmci(tau_max=tau_max, pc_alpha=None) q_matrix = pcmci.get_corrected_pvalues(p_matrix=results['p_matrix'], fdr_method='fdr_bh') link_matrix = pcmci.return_significant_links(pq_matrix=q_matrix, val_matrix=results['val_matrix'], alpha_level=0.01)['link_matrix'] if plot_res: tp.plot_graph(val_matrix=results['val_matrix'], link_matrix=link_matrix, var_names=var_names, link_colorbar_label='cross-MCI', node_colorbar_label='auto-MCI') plt.show() return link_matrix
def plot_results(pcmci, results, cond_ind_test, pc_alpha, tau_min, tau_max,
var_names):
"""Generates a network and a timeseries plot for the given results of the PCMCI algorithm and saves them as files.
Args:
pcmci: The PCMCI object used.
results: The results of the PCMCI algorithm.
cond_ind_test: The conditional independence test used.
pc_alpha: Alpha threshold for significance of links.
tau_min: The minimum lag.
tau_max: The maximum lag.
var_names: The names of the variables in the data.
"""
base_path = "experiments/causal_discovery/results/"
q_matrix = pcmci.get_corrected_pvalues(p_matrix=results['p_matrix'],
fdr_method='fdr_bh')
link_matrix = pcmci.return_significant_parents(
pq_matrix=q_matrix, val_matrix=results['val_matrix'],
alpha_level=0.01)['link_matrix']
class_prefix = ""
if isinstance(cond_ind_test, RCOT):
class_prefix = f"{cond_ind_test.measure}_num_f_{cond_ind_test.num_f}"
else:
class_prefix = cond_ind_test.measure
file_name_prefix = f"{class_prefix}_alpha_{pc_alpha:.5f}_tau_{tau_min}to{tau_max}"
tp.plot_graph(val_matrix=results['val_matrix'],
link_matrix=link_matrix,
var_names=var_names,
link_colorbar_label='cross-MCI',
node_colorbar_label='auto-MCI',
figsize=(20, 20),
label_fontsize=24,
node_label_size=24,
link_label_fontsize=14,
save_name=base_path + file_name_prefix + "_graph.png")
tp.plot_time_series_graph(val_matrix=results['val_matrix'],
link_matrix=link_matrix,
var_names=var_names,
link_colorbar_label='MCI',
figsize=(20, 20),
label_fontsize=24,
node_label_size=24,
save_name=base_path + file_name_prefix +
"_time_series_graph.png")
q_matrix = pcmci.get_corrected_pvalues(p_matrix=results['p_matrix'],
fdr_method='fdr_bh')
pcmci.print_significant_links(p_matrix=results['p_matrix'],
q_matrix=q_matrix,
val_matrix=results['val_matrix'],
alpha_level=0.01)
link_matrix = pcmci.return_significant_parents(
pq_matrix=q_matrix, val_matrix=results['val_matrix'],
alpha_level=0.01)['link_matrix']
graph = tp.plot_graph(
val_matrix=results['val_matrix'],
link_matrix=link_matrix,
var_names=headers,
link_colorbar_label='cross-MCI',
node_colorbar_label='auto-MCI',
)
if verbose > 1:
if display_images:
graph[0].show()
if save_images:
graph[0].savefig("causal.png")
if verbose > 0:
# Plot time series graph
if display_images:
plot = tp.plot_time_series_graph(val_matrix=results['val_matrix'],
link_matrix=link_matrix,
data_mask = np.zeros(data.shape)
for t in range(1, T + 1):
if (t % 73) >= 12 and (t % 73) <= 30:
data_mask[t - 1, :] = True
# Initialize dataframe object, specify time axis and variable names
var_names = ['WPSH', 'IO', 'WNP', 'ENSO', 'NAO']
dataframe = pp.DataFrame(data, mask=data_mask)
parcorr = ParCorr(significance='analytic', mask_type='xyz')
pcmci = PCMCI(dataframe=dataframe, cond_ind_test=parcorr)
results = pcmci.run_pcmci(tau_max=12, pc_alpha=0.03)
# Correct p-values
q_matrix = pcmci.get_corrected_pvalues(p_matrix=results['p_matrix'],
fdr_method='fdr_bh')
# Plotting
link_matrix = pcmci.return_significant_parents(
pq_matrix=q_matrix, val_matrix=results['val_matrix'],
alpha_level=0.03)['link_matrix']
tp.plot_graph(val_matrix=results['val_matrix'],
link_matrix=link_matrix,
var_names=var_names)
"""
tp.plot_time_series_graph(val_matrix=results['val_matrix'],
link_matrix=link_matrix, var_names=var_names)
"""
plt.savefig('./NWPAC.svg')
# In[11]:
link_matrix = pcmci.return_significant_links(pq_matrix=q_matrix,
val_matrix=results['val_matrix'],
alpha_level=0.01)['link_matrix']
# In[12]:
link_matrix.shape
# In[13]:
tp.plot_graph(
figsize=(5, 5),
val_matrix=results['val_matrix'],
link_matrix=link_matrix,
var_names=var_names,
link_colorbar_label='cross-MCI',
node_colorbar_label='auto-MCI',
)
plt.show()
# In[14]:
# Plot time series graph
tp.plot_time_series_graph(
val_matrix=results['val_matrix'],
link_matrix=link_matrix,
var_names=var_names,
link_colorbar_label='MCI',
)
plt.show()
results = pcmci.run_pcmci(tau_min = min_lag, tau_max=max_lag, pc_alpha=None)
#
q_matrix = pcmci.get_corrected_pvalues(p_matrix=results['p_matrix'], fdr_method='fdr_bh')
#
pcmci.print_significant_links(
p_matrix = results['p_matrix'],
q_matrix = q_matrix,
val_matrix = results['val_matrix'],
alpha_level = 0.05)
link_matrix = pcmci.return_significant_links(pq_matrix = results['p_matrix'],
val_matrix=results['val_matrix'], alpha_level=0.05)['link_matrix']
tp.plot_graph(
val_matrix=results['val_matrix'],
link_matrix=link_matrix,
var_names=study_data.columns,
link_colorbar_label='cross-MCI',
node_colorbar_label='auto-MCI',
)
plt.show()
#dataframe = dataframe.iloc[:,:-4]
G = nx.DiGraph()
for i , node_name in enumerate(dataframe.var_names):
# G.add_node((i,{'name':node_name}))
G.add_node(i,name = node_name, influenced_by = 0)
#make N^2 matrix
n_connections = 0
all_tau_link = np.max(link_matrix, axis = 2)
if graph_pcmci[j, p[0], 0] == "o-o":
Unoriented = 1
# "unoriented link"
elif graph_pcmci[p[0], j, abs(p[1])] == "x-x":
Unoriented = 1
# "unclear orientation due to conflict"
links_df = links_df.append(pd.DataFrame({
'VarSou': [VarSou],
'VarTar': [VarTar],
'Source': [Source],
'Target': [Target],
'TimeLag': [TimeLag],
'Strength': [Strength],
'Unoriented': [Unoriented]
}),
ignore_index=True)
# remove the self correlation edges
links_df = links_df.loc[links_df['Source'] != links_df['Target']]
print(links_df.sort_values(by=['Source', 'Target']))
links_df.to_csv('C:\\Users\\thril\\Desktop\\123.csv')
# print(graph_pcmci)
tp.plot_graph(
val_matrix=results_pcmci['val_matrix'],
link_matrix=results_pcmci['graph'],
var_names=p_agent_names,
link_colorbar_label='cross-MCI (edges)',
node_colorbar_label='auto-MCI (nodes)',
)
plt.show()