def test_plot_box(self):
"""测试可视化代码"""
# plot box，使用seaborn库
keys = ["Bias", "NSE", "FLV", "FHV"]
inds_test = subset_of_dict(self.inds, keys)
plot_diff_boxes(inds_test,
title_str="Metrics of streamflow prediction")
plt.savefig(os.path.join(self.dir_out, 'boxes.png'),
dpi=500,
bbox_inches="tight")
plt.show()
def test_ensemble_results_plot_box(self):
preds = []
obss = []
# cases_exps = ["basic_exp11", "basic_exp17"]
cases_exps = [
"basic_exp12", "basic_exp13", "basic_exp14", "basic_exp15",
"basic_exp16", "basic_exp18"
]
for case_exp in cases_exps:
config_data_i = load_dataconfig_case_exp(case_exp)
pred_i, obs_i = load_result(config_data_i.data_path['Temp'],
self.test_epoch)
pred_i = pred_i.reshape(pred_i.shape[0], pred_i.shape[1])
obs_i = obs_i.reshape(obs_i.shape[0], obs_i.shape[1])
print(obs_i)
preds.append(pred_i)
obss.append(obs_i)
preds_np = np.array(preds)
obss_np = np.array(obss)
pred_mean = np.mean(preds_np, axis=0)
obs_mean = np.mean(obss_np, axis=0)
inds = statError(obs_mean, pred_mean)
keys = ["Bias", "RMSE", "NSE"]
inds_test = subset_of_dict(inds, keys)
box_fig = plot_diff_boxes(inds_test)
def plot_we_need(data_model_test,
obs,
pred,
show_me_num=5,
point_file=None,
**kwargs):
pred = pred.reshape(pred.shape[0], pred.shape[1])
obs = obs.reshape(pred.shape[0], pred.shape[1])
inds = statError(obs, pred)
# plot box
keys = ["Bias", "RMSE", "NSE"]
inds_test = subset_of_dict(inds, keys)
box_fig = plot_diff_boxes(inds_test)
box_fig.savefig(
os.path.join(data_model_test.data_source.data_config.data_path["Out"],
"box_fig.png"))
# plot ts
t_s_dict = data_model_test.t_s_dict
sites = np.array(t_s_dict["sites_id"])
t_range = np.array(t_s_dict["t_final_range"])
ts_fig = plot_ts_obs_pred(obs, pred, sites, t_range, show_me_num)
ts_fig.savefig(
os.path.join(data_model_test.data_source.data_config.data_path["Out"],
"ts_fig.png"))
# plot nse ecdf
sites_df_nse = pd.DataFrame({"sites": sites, keys[2]: inds_test[keys[2]]})
plot_ecdf(sites_df_nse, keys[2])
# plot map
if point_file is None:
gauge_dict = data_model_test.data_source.gage_dict
plot_map(gauge_dict, sites_df_nse, **kwargs)
else:
plot_ind_map(point_file, sites_df_nse, percentile=25)
def test_inv_plot(self):
data_model = GagesModel.load_datamodel(
self.config_data_2.data_path["Temp"],
"2",
data_source_file_name='test_data_source.txt',
stat_file_name='test_Statistics.json',
flow_file_name='test_flow.npy',
forcing_file_name='test_forcing.npy',
attr_file_name='test_attr.npy',
f_dict_file_name='test_dictFactorize.json',
var_dict_file_name='test_dictAttribute.json',
t_s_dict_file_name='test_dictTimeSpace.json')
flow_pred_file = os.path.join(
data_model.data_source.data_config.data_path['Temp'],
'epoch' + str(self.test_epoch) + 'flow_pred.npy')
flow_obs_file = os.path.join(
data_model.data_source.data_config.data_path['Temp'],
'epoch' + str(self.test_epoch) + 'flow_obs.npy')
pred = unserialize_numpy(flow_pred_file)
obs = unserialize_numpy(flow_obs_file)
pred = pred.reshape(pred.shape[0], pred.shape[1])
obs = obs.reshape(obs.shape[0], obs.shape[1])
inds = statError(obs, pred)
inds['STAID'] = data_model.t_s_dict["sites_id"]
inds_df = pd.DataFrame(inds)
inds_df.to_csv(
os.path.join(self.config_data_2.data_path["Out"], 'data_df.csv'))
# plot box，使用seaborn库
keys = ["Bias", "RMSE", "NSE"]
inds_test = subset_of_dict(inds, keys)
box_fig = plot_diff_boxes(inds_test)
box_fig.savefig(
os.path.join(self.config_data_2.data_path["Out"], "box_fig.png"))
# plot ts
show_me_num = 5
t_s_dict = data_model.t_s_dict
sites = np.array(t_s_dict["sites_id"])
t_range = np.array(t_s_dict["t_final_range"])
time_seq_length = self.config_data_1.model_dict['model']['seqLength']
time_start = np.datetime64(t_range[0]) + np.timedelta64(
time_seq_length - 1, 'D')
t_range[0] = np.datetime_as_string(time_start, unit='D')
ts_fig = plot_ts_obs_pred(obs, pred, sites, t_range, show_me_num)
ts_fig.savefig(
os.path.join(self.config_data_2.data_path["Out"], "ts_fig.png"))
# plot nse ecdf
sites_df_nse = pd.DataFrame({
"sites": sites,
keys[2]: inds_test[keys[2]]
})
plot_ecdf(sites_df_nse, keys[2])
# plot map
gauge_dict = data_model.data_source.gage_dict
plot_map(gauge_dict,
sites_df_nse,
id_col="STAID",
lon_col="LNG_GAGE",
lat_col="LAT_GAGE")
def test_forecast_test(self):
sim_df = GagesModel.load_datamodel(
self.sim_config_data.data_path["Temp"],
"1",
data_source_file_name='test_data_source.txt',
stat_file_name='test_Statistics.json',
flow_file_name='test_flow.npy',
forcing_file_name='test_forcing.npy',
attr_file_name='test_attr.npy',
f_dict_file_name='test_dictFactorize.json',
var_dict_file_name='test_dictAttribute.json',
t_s_dict_file_name='test_dictTimeSpace.json')
df = GagesModel.load_datamodel(
self.config_data.data_path["Temp"],
"2",
data_source_file_name='test_data_source.txt',
stat_file_name='test_Statistics.json',
flow_file_name='test_flow.npy',
forcing_file_name='test_forcing.npy',
attr_file_name='test_attr.npy',
f_dict_file_name='test_dictFactorize.json',
var_dict_file_name='test_dictAttribute.json',
t_s_dict_file_name='test_dictTimeSpace.json')
data_input = GagesForecastDataModel(sim_df, df)
pred, obs = test_lstm_forecast(data_input)
pred = pred.reshape(pred.shape[0], pred.shape[1])
obs = obs.reshape(obs.shape[0], obs.shape[1])
inds = statError(obs, pred)
show_me_num = 5
t_s_dict = data_input.model_data.t_s_dict
sites = np.array(t_s_dict["sites_id"])
t_range = np.array(t_s_dict["t_final_range"])
time_seq_length = data_input.model_data.data_source.data_config.model_dict[
'model']['seqLength']
time_start = np.datetime64(t_range[0]) + np.timedelta64(
time_seq_length, 'D')
t_range[0] = np.datetime_as_string(time_start, unit='D')
ts_fig = plot_ts_obs_pred(obs, pred, sites, t_range, show_me_num)
ts_fig.savefig(
os.path.join(self.config_data.data_path["Out"], "ts_fig.png"))
# # plot box，使用seaborn库
keys = ["Bias", "RMSE", "NSE"]
inds_test = subset_of_dict(inds, keys)
box_fig = plot_diff_boxes(inds_test)
box_fig.savefig(
os.path.join(self.config_data.data_path["Out"], "box_fig.png"))
# plot map
sites_df = pd.DataFrame({"sites": sites, keys[2]: inds_test[keys[2]]})
plot_ind_map(df.data_source.all_configs['gage_point_file'], sites_df)
def test_explore_test(self):
models_num = 0
dirs = os.listdir(self.config_data.data_path["Temp"])
for dir_temp in dirs:
if os.path.isdir(
os.path.join(self.config_data.data_path["Temp"],
dir_temp)):
models_num += 1
for count in range(models_num):
print("\n", "testing model", str(count + 1), ":\n")
data_model = GagesModel.load_datamodel(
self.config_data.data_path["Temp"],
str(count),
data_source_file_name='test_data_source.txt',
stat_file_name='test_Statistics.json',
flow_file_name='test_flow.npy',
forcing_file_name='test_forcing.npy',
attr_file_name='test_attr.npy',
f_dict_file_name='test_dictFactorize.json',
var_dict_file_name='test_dictAttribute.json',
t_s_dict_file_name='test_dictTimeSpace.json')
pred, obs = master_test(data_model)
pred = pred.reshape(pred.shape[0], pred.shape[1])
obs = obs.reshape(obs.shape[0], obs.shape[1])
inds = statError(obs, pred)
show_me_num = 5
t_s_dict = data_model.t_s_dict
sites = np.array(t_s_dict["sites_id"])
t_range = np.array(t_s_dict["t_final_range"])
ts_fig = plot_ts_obs_pred(obs, pred, sites, t_range, show_me_num)
ts_fig.savefig(
os.path.join(
data_model.data_source.data_config.data_path["Out"],
"ts_fig.png"))
# # plot box，使用seaborn库
keys = ["Bias", "RMSE", "NSE"]
inds_test = subset_of_dict(inds, keys)
box_fig = plot_diff_boxes(inds_test)
box_fig.savefig(
os.path.join(
data_model.data_source.data_config.data_path["Out"],
"box_fig.png"))
# plot map
sites_df = pd.DataFrame({
"sites": sites,
keys[2]: inds_test[keys[2]]
})
plot_ind_map(data_model.data_source.all_configs['gage_point_file'],
sites_df)
def test_plot_pretrained_model_test(self):
data_model_test = GagesModel.load_datamodel(
self.majordam_config_data.data_path["Temp"],
data_source_file_name='test_data_source.txt',
stat_file_name='test_Statistics.json',
flow_file_name='test_flow.npy',
forcing_file_name='test_forcing.npy',
attr_file_name='test_attr.npy',
f_dict_file_name='test_dictFactorize.json',
var_dict_file_name='test_dictAttribute.json',
t_s_dict_file_name='test_dictTimeSpace.json')
pretrained_model_name = "nodam" + self.nomajordam_subdir.split(
"/")[1] + "_pretrained_model"
save_dir = os.path.join(
data_model_test.data_source.data_config.data_path['Out'],
pretrained_model_name)
pred, obs = load_result(save_dir, self.test_epoch)
pred = pred.reshape(pred.shape[0], pred.shape[1])
obs = obs.reshape(pred.shape[0], pred.shape[1])
inds = statError(obs, pred)
# plot box，使用seaborn库
keys = ["Bias", "RMSE", "NSE"]
inds_test = subset_of_dict(inds, keys)
box_fig = plot_diff_boxes(inds_test)
box_fig.savefig(os.path.join(save_dir, "box_fig.png"))
# plot ts
show_me_num = 5
t_s_dict = data_model_test.t_s_dict
sites = np.array(t_s_dict["sites_id"])
t_range = np.array(t_s_dict["t_final_range"])
ts_fig = plot_ts_obs_pred(obs, pred, sites, t_range, show_me_num)
ts_fig.savefig(os.path.join(save_dir, "ts_fig.png"))
# plot nse ecdf
sites_df_nse = pd.DataFrame({
"sites": sites,
keys[2]: inds_test[keys[2]]
})
plot_ecdf(sites_df_nse, keys[2])
# plot map
gauge_dict = data_model_test.data_source.gage_dict
plot_map(gauge_dict,
sites_df_nse,
id_col="STAID",
lon_col="LNG_GAGE",
lat_col="LAT_GAGE")
def plot_region_seperately(gages_data_model, epoch, id_regions_idx, preds,
obss, inds_dfs):
df_id_region = np.array(gages_data_model.t_s_dict["sites_id"])
regions_name = gages_data_model.data_source.all_configs.get("regions")
for i in range(len(id_regions_idx)):
# plot box
keys = ["Bias", "RMSE", "NSE"]
inds_test = subset_of_dict(inds_dfs[i], keys)
box_fig = plot_diff_boxes(inds_test)
box_fig.savefig(
os.path.join(
gages_data_model.data_source.data_config.data_path["Out"],
regions_name[i] + "epoch" + str(epoch) + "box_fig.png"))
# plot ts
sites = np.array(df_id_region[id_regions_idx[i]])
t_range = np.array(gages_data_model.t_s_dict["t_final_range"])
show_me_num = 5
ts_fig = plot_ts_obs_pred(obss[i], preds[i], sites, t_range,
show_me_num)
ts_fig.savefig(
os.path.join(
gages_data_model.data_source.data_config.data_path["Out"],
regions_name[i] + "epoch" + str(epoch) + "ts_fig.png"))
# plot nse ecdf
sites_df_nse = pd.DataFrame({
"sites": sites,
keys[2]: inds_test[keys[2]]
})
plot_ecdf(
sites_df_nse, keys[2],
os.path.join(
gages_data_model.data_source.data_config.data_path["Out"],
regions_name[i] + "epoch" + str(epoch) + "ecdf_fig.png"))
# plot map
gauge_dict = gages_data_model.data_source.gage_dict
save_map_file = os.path.join(
gages_data_model.data_source.data_config.data_path["Out"],
regions_name[i] + "epoch" + str(epoch) + "map_fig.png")
plot_map(gauge_dict,
sites_df_nse,
save_file=save_map_file,
id_col="STAID",
lon_col="LNG_GAGE",
lat_col="LAT_GAGE")
def test_plot_pretrained_model_test(self):
data_model_test = GagesModel.load_datamodel(
self.config_data.data_path["Temp"],
data_source_file_name='test_data_source.txt',
stat_file_name='test_Statistics.json',
flow_file_name='test_flow.npy',
forcing_file_name='test_forcing.npy',
attr_file_name='test_attr.npy',
f_dict_file_name='test_dictFactorize.json',
var_dict_file_name='test_dictAttribute.json',
t_s_dict_file_name='test_dictTimeSpace.json')
save_dir = data_model_test.data_source.data_config.data_path['Temp']
pred, obs = load_result(save_dir,
self.config_data.config_file.TEST_EPOCH)
pred = pred.reshape(pred.shape[0], pred.shape[1])
obs = obs.reshape(pred.shape[0], pred.shape[1])
inds = statError(obs, pred)
keys = ["Bias", "RMSE", "NSE"]
t_s_dict = data_model_test.t_s_dict
sites = np.array(t_s_dict["sites_id"])
t_range = np.array(t_s_dict["t_final_range"])
# plot nse ecdf
sites_df_nse = pd.DataFrame({"sites": sites, keys[2]: inds[keys[2]]})
plot_ecdf(sites_df_nse, keys[2])
# plot map
gauge_dict = data_model_test.data_source.gage_dict
plot_map(gauge_dict,
sites_df_nse,
id_col="STAID",
lon_col="LNG_GAGE",
lat_col="LAT_GAGE")
# plot box，使用seaborn库
inds_test = subset_of_dict(inds, keys)
box_fig = plot_diff_boxes(inds_test)
# plot ts
show_me_num = 5
ts_fig = plot_ts_obs_pred(obs, pred, sites, t_range, show_me_num)
plt.show()
def plot_gages_attrs_boxes(sites1,
sites2,
attr_lst,
attrs1,
attrs2,
diff_str,
row_and_col=None):
"""plot boxplots of GAGES model results"""
if type(sites1) == list or type(sites2) == list:
sites1 = np.array(sites1)
sites2 = np.array(sites2)
sites1diff = np.tile(0, sites1.size).reshape(sites1.size, 1)
site2diff = np.tile(1, sites2.size).reshape(sites2.size, 1)
attrs1_new = np.concatenate((attrs1, sites1diff), axis=1)
attrs2_new = np.concatenate((attrs2, site2diff), axis=1)
diff_str_lst = [diff_str]
df1 = pd.DataFrame(attrs1_new, columns=attr_lst + diff_str_lst)
df2 = pd.DataFrame(attrs2_new, columns=attr_lst + diff_str_lst)
result = pd.concat([df1, df2])
box_fig = plot_diff_boxes(result,
row_and_col=row_and_col,
y_col=np.arange(len(attr_lst)).tolist(),
x_col=len(attr_lst))
return box_fig
def test_purposes_seperate(self):
quick_data_dir = os.path.join(self.config_data.data_path["DB"],
"quickdata")
data_dir = os.path.join(quick_data_dir,
"allnonref-dam_95-05_nan-0.1_00-1.0")
data_model_test = GagesModel.load_datamodel(
data_dir,
data_source_file_name='test_data_source.txt',
stat_file_name='test_Statistics.json',
flow_file_name='test_flow.npy',
forcing_file_name='test_forcing.npy',
attr_file_name='test_attr.npy',
f_dict_file_name='test_dictFactorize.json',
var_dict_file_name='test_dictAttribute.json',
t_s_dict_file_name='test_dictTimeSpace.json')
data_model = GagesModel.update_data_model(self.config_data,
data_model_test)
nid_dir = os.path.join(
"/".join(self.config_data.data_path["DB"].split("/")[:-1]), "nid",
"quickdata")
gage_main_dam_purpose = unserialize_json(
os.path.join(nid_dir, "dam_main_purpose_dict.json"))
gage_main_dam_purpose_lst = list(gage_main_dam_purpose.values())
gage_main_dam_purpose_unique = np.unique(gage_main_dam_purpose_lst)
purpose_regions = {}
for i in range(gage_main_dam_purpose_unique.size):
sites_id = []
for key, value in gage_main_dam_purpose.items():
if value == gage_main_dam_purpose_unique[i]:
sites_id.append(key)
assert (all(x < y for x, y in zip(sites_id, sites_id[1:])))
purpose_regions[gage_main_dam_purpose_unique[i]] = sites_id
id_regions_idx = []
id_regions_sites_ids = []
df_id_region = np.array(data_model.t_s_dict["sites_id"])
for key, value in purpose_regions.items():
gages_id = value
c, ind1, ind2 = np.intersect1d(df_id_region,
gages_id,
return_indices=True)
assert (all(x < y for x, y in zip(ind1, ind1[1:])))
assert (all(x < y for x, y in zip(c, c[1:])))
id_regions_idx.append(ind1)
id_regions_sites_ids.append(c)
pred_all, obs_all = load_result(self.config_data.data_path["Temp"],
self.test_epoch)
pred_all = pred_all.reshape(pred_all.shape[0], pred_all.shape[1])
obs_all = obs_all.reshape(obs_all.shape[0], obs_all.shape[1])
for i in range(9, len(gage_main_dam_purpose_unique)):
pred = pred_all[id_regions_idx[i], :]
obs = obs_all[id_regions_idx[i], :]
inds = statError(obs, pred)
inds['STAID'] = id_regions_sites_ids[i]
inds_df = pd.DataFrame(inds)
inds_df.to_csv(
os.path.join(
self.config_data.data_path["Out"],
gage_main_dam_purpose_unique[i] + "epoch" +
str(self.test_epoch) + 'data_df.csv'))
# plot box，使用seaborn库
keys = ["Bias", "RMSE", "NSE"]
inds_test = subset_of_dict(inds, keys)
box_fig = plot_diff_boxes(inds_test)
box_fig.savefig(
os.path.join(
self.config_data.data_path["Out"],
gage_main_dam_purpose_unique[i] + "epoch" +
str(self.test_epoch) + "box_fig.png"))
# plot ts
sites = np.array(df_id_region[id_regions_idx[i]])
t_range = np.array(data_model.t_s_dict["t_final_range"])
show_me_num = 1
ts_fig = plot_ts_obs_pred(obs, pred, sites, t_range, show_me_num)
ts_fig.savefig(
os.path.join(
self.config_data.data_path["Out"],
gage_main_dam_purpose_unique[i] + "epoch" +
str(self.test_epoch) + "ts_fig.png"))
# plot nse ecdf
sites_df_nse = pd.DataFrame({
"sites": sites,
keys[2]: inds_test[keys[2]]
})
plot_ecdf(
sites_df_nse, keys[2],
os.path.join(
self.config_data.data_path["Out"],
gage_main_dam_purpose_unique[i] + "epoch" +
str(self.test_epoch) + "ecdf_fig.png"))
# plot map
gauge_dict = data_model.data_source.gage_dict
save_map_file = os.path.join(
self.config_data.data_path["Out"],
gage_main_dam_purpose_unique[i] + "epoch" +
str(self.test_epoch) + "map_fig.png")
plot_map(gauge_dict,
sites_df_nse,
save_file=save_map_file,
id_col="STAID",
lon_col="LNG_GAGE",
lat_col="LAT_GAGE")
inds_df[show_ind_NSE] <= nse_range[1])].index.tolist()
plot_gages_map(data_model,
inds_df,
show_ind_NSE,
idx_lstl_nse,
cbar_font_size=14)
plt.savefig(os.path.join(config_data.data_path["Out"], 'map_NSE.png'),
dpi=FIGURE_DPI,
bbox_inches="tight")
# plt.figure()
# plot box，使用seaborn库
keys = ["Bias", "NSE", "FHV", "FLV"]
inds_test = subset_of_dict(inds_df, keys)
plot_diff_boxes(inds_test)
plt.savefig(os.path.join(config_data.data_path["Out"], 'boxes.png'),
dpi=FIGURE_DPI,
bbox_inches="tight")
# plt.figure()
############################ plot map box ###########################
# plot NSE
nse_range = [0, 1]
# nse_range = [-10000, 1]
# nse_range = [-10000, 0]
idx_lstl_nse = inds_df[(inds_df[show_ind_NSE] >= nse_range[0]) & (
inds_df[show_ind_NSE] <= nse_range[1])].index.tolist()
plot_gages_map_and_box(data_model,
inds_df,
show_ind_NSE,