def call_rnn(self, other_args: List[str]):
"""Process rnn command"""
try:
ns_parser = pred_helper.parse_args(
prog="rnn",
description="""Recurrent Neural Network. """,
other_args=other_args,
)
if ns_parser:
neural_networks_view.display_rnn(
dataset=self.coin,
data=self.data[self.target],
n_input_days=ns_parser.n_inputs,
n_predict_days=ns_parser.n_days,
learning_rate=ns_parser.lr,
epochs=ns_parser.n_epochs,
batch_size=ns_parser.n_batch_size,
test_size=ns_parser.valid_split,
n_loops=ns_parser.n_loops,
no_shuffle=ns_parser.no_shuffle,
time_res=self.resolution,
)
except Exception as e:
console.print(e, "\n")
finally:
pred_helper.restore_env()
def call_rnn(self, other_args: List[str]):
"""Process rnn command"""
try:
ns_parser = pred_helper.parse_args(
prog="rnn",
description="""Recurrent Neural Network. """,
other_args=other_args,
)
if ns_parser:
if ns_parser.n_inputs > len(self.data):
console.print(
f"[red]Data only contains {len(self.data)} samples and the model is trying "
f"to use {ns_parser.n_inputs} inputs. Either use less inputs or load with"
f" an earlier start date[/red]\n")
return
neural_networks_view.display_rnn(
dataset=self.current_id,
data=self.data,
n_input_days=ns_parser.n_inputs,
n_predict_days=ns_parser.n_days,
learning_rate=ns_parser.lr,
epochs=ns_parser.n_epochs,
batch_size=ns_parser.n_batch_size,
test_size=ns_parser.valid_split,
n_loops=ns_parser.n_loops,
no_shuffle=ns_parser.no_shuffle,
time_res=self.resolution,
)
except Exception as e:
console.print(e)
finally:
pred_helper.restore_env()