for tmp in load_dataset(config.set_params["path"], subset="val")
]
# ------ save hyperparameters -------
os.makedirs(train_params["save_path"][-1], exist_ok=True)
with open(os.path.join(train_params["save_path"][-1], "hyperparameters.txt"),
"w") as file:
for key, value in subnet_params.items():
file.write(key + ": " + str(value) + "\n")
for key, value in it_net_params.items():
file.write(key + ": " + str(value) + "\n")
for key, value in train_params.items():
file.write(key + ": " + str(value) + "\n")
file.write("train_phases" + ": " + str(train_phases) + "\n")
# ------ construct network and train -----
subnet = subnet(**subnet_params).to(device)
it_net = IterativeNet(subnet, **it_net_params).to(device)
for i in range(train_phases):
train_params_cur = {}
for key, value in train_params.items():
train_params_cur[key] = (value[i] if isinstance(value,
(tuple,
list)) else value)
print("Phase {}:".format(i + 1))
for key, value in train_params_cur.items():
print(key + ": " + str(value))
it_net.train_on((Y_train, X_train), (Y_val, X_val), **train_params_cur)
for key, value in subnet_params.items():
file.write(key + ": " + str(value) + "\n")
for key, value in it_net_params.items():
file.write(key + ": " + str(value) + "\n")
for key, value in train_params.items():
file.write(key + ": " + str(value) + "\n")
for key, value in train_data_params.items():
file.write(key + ": " + str(value) + "\n")
for key, value in val_data_params.items():
file.write(key + ": " + str(value) + "\n")
file.write("train_phases" + ": " + str(train_phases) + "\n")
# ------ construct network and train -----
subnet = subnet(**subnet_params).to(device)
it_net = IterativeNet(subnet, **it_net_params).to(device)
train_data = train_data("train", **train_data_params)
val_data = val_data("val", **val_data_params)
for i in range(train_phases):
train_params_cur = {}
for key, value in train_params.items():
train_params_cur[key] = (value[i] if isinstance(value,
(tuple,
list)) else value)
print("Phase {}:".format(i + 1))
for key, value in train_params_cur.items():
print(key + ": " + str(value))
it_net.train_on(train_data, val_data, **train_params_cur)