def test_train_camels(self):
data_model = GagesModel.load_datamodel(self.config_data.data_path["Temp"],
data_source_file_name='data_source.txt',
stat_file_name='Statistics.json', flow_file_name='flow.npy',
forcing_file_name='forcing.npy', attr_file_name='attr.npy',
f_dict_file_name='dictFactorize.json',
var_dict_file_name='dictAttribute.json',
t_s_dict_file_name='dictTimeSpace.json')
with torch.cuda.device(1):
data_models = GagesModel.every_model(data_model)
for i in range(len(data_models)):
print("\n", "Training model", str(i + 1), ":\n")
model, train_loss, valid_loss = master_train(data_models[i], valid_size=0.2)
fig = plot_loss_early_stop(train_loss, valid_loss)
out_dir = self.config_data.data_path["Out"]
fig.savefig(os.path.join(out_dir, 'loss_plot.png'), bbox_inches='tight')
def test_train_gages_iter(self):
data_model = GagesModel.load_datamodel(self.config_data.data_path["Temp"],
data_source_file_name='data_source.txt',
stat_file_name='Statistics.json', flow_file_name='flow.npy',
forcing_file_name='forcing.npy', attr_file_name='attr.npy',
f_dict_file_name='dictFactorize.json',
var_dict_file_name='dictAttribute.json',
t_s_dict_file_name='dictTimeSpace.json')
valid_size = 0.2
with torch.cuda.device(1):
for i in range(145, data_model.data_flow.shape[0]):
print("\n", "Training model", str(i + 1), ":\n")
data_models_i = GagesModel.which_data_model(data_model, i)
model, train_loss, valid_loss = master_train_1by1(data_models_i, valid_size=valid_size)
fig = plot_loss_early_stop(train_loss, valid_loss)
out_dir = data_models_i.data_source.data_config.data_path["Out"]
fig.savefig(os.path.join(out_dir, 'loss_plot.png'), bbox_inches='tight')
def test_train_valid_camels(self):
data_model = CamelsModel.load_datamodel(
self.config_data.data_path["Temp"],
data_source_file_name='data_source.txt',
stat_file_name='Statistics.json',
flow_file_name='flow.npy',
forcing_file_name='forcing.npy',
attr_file_name='attr.npy',
f_dict_file_name='dictFactorize.json',
var_dict_file_name='dictAttribute.json',
t_s_dict_file_name='dictTimeSpace.json')
with torch.cuda.device(2):
model, train_loss, valid_loss = master_train(data_model,
valid_size=0.2)
fig = plot_loss_early_stop(train_loss, valid_loss)
out_dir = self.config_data.data_path["Out"]
fig.savefig(os.path.join(out_dir, 'loss_plot.png'),
bbox_inches='tight')
