jerry/OpenBB
1
0
Fork 0
mirror of https://github.com/OpenBB-finance/OpenBB.git synced 2026-10-28 04:52:07 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merge pull request #4178 from OpenBB-finance/feature/forecast-metrics

Browse Source 
[FEAT] Additional metrics for evaluating forecasting (RMSE+ MSE)
This commit is contained in:
James Maslek
2023-02-09 15:03:28 -05:00
committed by GitHub
parent 6e484e8256 9895cb2e48
commit 583bb1df05
26 changed files with 214 additions and 18 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
.gitignore vendored
View File

@@ -61,3 +61,6 @@ darts_logs/
# User data
custom_imports/*.csv
custom_imports/*/*.csv
# lightning logs
lightning_logs/

2
build/conda/conda-3-9-env-full.yaml vendored
View File

@@ -5,4 +5,4 @@ dependencies:
- python=3.9.6
- pip
- poetry=1.1.13
- lightgbm=3.3.3
- lightgbm=3.3.5

4
openbb_terminal/forecast/brnn_model.py
View File

@@ -38,6 +38,7 @@ def get_brnn_data(
model_save_name: str = "brnn_model",
force_reset: bool = True,
save_checkpoints: bool = True,
metric: str = "mape",
) -> Tuple[
Optional[List[TimeSeries]],
Optional[List[TimeSeries]],
@@ -84,6 +85,8 @@ def get_brnn_data(
discarded). Defaults to True.
save_checkpoints: bool
Whether or not to automatically save the untrained model and checkpoints from training. Defaults to True.
metric: str
Metric to use for model selection. Defaults to "mape".
Returns
-------
@@ -174,4 +177,5 @@ def get_brnn_data(
train_split,
forecast_horizon,
n_predict,
metric,
)

7
openbb_terminal/forecast/brnn_view.py
View File

@@ -43,6 +43,7 @@ def display_brnn_forecast(
end_date: Optional[datetime] = None,
naive: bool = False,
export_pred_raw: bool = False,
metric: str = "mape",
external_axes: Optional[List[plt.axes]] = None,
):
"""Display BRNN forecast
@@ -102,6 +103,10 @@ def display_brnn_forecast(
naive: bool
Whether to show the naive baseline. This just assumes the closing price will be the same
as the previous day's closing price. Defaults to False.
export_pred_raw: bool
Whether to export the raw predicted values. Defaults to False.
metric: str
The metric to use for the model. Defaults to "mape".
external_axes: Optional[List[plt.axes]]
External axes to plot on
"""
@@ -137,6 +142,7 @@ def display_brnn_forecast(
model_save_name=model_save_name,
force_reset=force_reset,
save_checkpoints=save_checkpoints,
metric=metric,
)
if ticker_series == []:
return
@@ -160,6 +166,7 @@ def display_brnn_forecast(
forecast_only=forecast_only,
naive=naive,
export_pred_raw=export_pred_raw,
metric=metric,
external_axes=external_axes,
)
if residuals:

20
openbb_terminal/forecast/expo_model.py
View File

@@ -8,7 +8,7 @@ from typing import List, Optional, Tuple, Union
import pandas as pd
from darts import TimeSeries
from darts.metrics import mape
from darts.metrics import mape, mse, rmse, smape
from darts.models import ExponentialSmoothing
from darts.utils.utils import ModelMode, SeasonalityMode
from numpy import ndarray
@@ -40,6 +40,7 @@ def get_expo_data(
n_predict: int = 5,
start_window: float = 0.85,
forecast_horizon: int = 5,
metric: str = "mape",
) -> Tuple[
List[TimeSeries],
List[TimeSeries],
@@ -76,6 +77,8 @@ def get_expo_data(
Size of sliding window from start of timeseries and onwards
forecast_horizon: int
Number of days to forecast when backtesting and retraining historical
metric: str
Metric to use for backtesting. Defaults to MAPE.
Returns
-------
@@ -149,8 +152,19 @@ def get_expo_data(
# we have the historical fcast, now lets train on entire set and predict.
best_model.fit(ticker_series)
probabilistic_forecast = best_model.predict(int(n_predict), num_samples=500)
precision = mape(actual_series=ticker_series, pred_series=historical_fcast_es)
console.print(f"Exponential smoothing obtains MAPE: {precision:.2f}% \n")
if metric == "rmse":
precision = rmse(actual_series=ticker_series, pred_series=historical_fcast_es)
elif metric == "mse":
precision = mse(actual_series=ticker_series, pred_series=historical_fcast_es)
elif metric == "mape":
precision = mape(actual_series=ticker_series, pred_series=historical_fcast_es)
elif metric == "smape":
precision = smape(actual_series=ticker_series, pred_series=historical_fcast_es)
console.print(
f"Exponential smoothing obtains {metric.upper()}: {precision:.2f}% \n"
)
return (
ticker_series,

7
openbb_terminal/forecast/expo_view.py
View File

@@ -36,6 +36,7 @@ def display_expo_forecast(
end_date: Optional[datetime] = None,
naive: bool = False,
export_pred_raw: bool = False,
metric: str = "mape",
external_axes: Optional[List[plt.axes]] = None,
):
"""Display Probabilistic Exponential Smoothing forecast
@@ -80,6 +81,10 @@ def display_expo_forecast(
naive: bool
Whether to show the naive baseline. This just assumes the closing price will be the same
as the previous day's closing price. Defaults to False.
export_pred_raw: bool
Whether to export the raw predicted values. Defaults to False.
metric: str
The metric to use when backtesting. Defaults to "mape".
external_axes: Optional[List[plt.axes]]
External axes to plot on
"""
@@ -103,6 +108,7 @@ def display_expo_forecast(
n_predict=n_predict,
start_window=start_window,
forecast_horizon=forecast_horizon,
metric=metric,
)
if ticker_series == []:
@@ -127,6 +133,7 @@ def display_expo_forecast(
forecast_only=forecast_only,
naive=naive,
export_pred_raw=export_pred_raw,
metric=metric,
external_axes=external_axes,
)
if residuals:

36
openbb_terminal/forecast/forecast_controller.py
View File

@@ -356,6 +356,7 @@ class ForecastController(BaseController):
naive: bool = False,
explainability_raw: bool = False,
export_pred_raw: bool = False,
metric: bool = False,
):
if hidden_size:
parser.add_argument(
@@ -531,7 +532,7 @@ class ForecastController(BaseController):
action="store",
dest="model_type",
default="LSTM",
help='Either a string specifying the RNN module type ("RNN", "LSTM" or "GRU")',
help='Enter a string specifying the RNN module type ("RNN", "LSTM" or "GRU")',
)
if dropout is not None:
parser.add_argument(
@@ -631,6 +632,17 @@ class ForecastController(BaseController):
help="Export predictions to a csv file.",
)
if metric:
parser.add_argument(
"--metric",
type=str,
action="store",
dest="metric",
default="mape",
choices=["rmse", "mse", "mape", "smape"],
help="Calculate precision based on a specific metric (rmse, mse, mape)",
)
# if user does not put in --dataset
return parser
@@ -2114,6 +2126,7 @@ class ForecastController(BaseController):
end=True,
naive=True,
export_pred_raw=True,
metric=True,
)
ns_parser = self.parse_known_args_and_warn(
parser,
@@ -2143,6 +2156,7 @@ class ForecastController(BaseController):
end_date=ns_parser.s_end_date,
naive=ns_parser.naive,
export_pred_raw=ns_parser.export_pred_raw,
metric=ns_parser.metric,
)
@log_start_end(log=logger)
@@ -2175,6 +2189,7 @@ class ForecastController(BaseController):
end=True,
naive=True,
export_pred_raw=True,
metric=True,
)
ns_parser = self.parse_known_args_and_warn(
parser,
@@ -2202,6 +2217,7 @@ class ForecastController(BaseController):
end_date=ns_parser.s_end_date,
naive=ns_parser.naive,
export_pred_raw=ns_parser.export_pred_raw,
metric=ns_parser.metric,
)
@log_start_end(log=logger)
@@ -2261,6 +2277,7 @@ class ForecastController(BaseController):
end=True,
naive=True,
export_pred_raw=True,
metric=True,
)
ns_parser = self.parse_known_args_and_warn(
parser,
@@ -2297,6 +2314,7 @@ class ForecastController(BaseController):
end_date=ns_parser.s_end_date,
naive=ns_parser.naive,
export_pred_raw=ns_parser.export_pred_raw,
metric=ns_parser.metric,
)
@log_start_end(log=logger)
@@ -2373,6 +2391,7 @@ class ForecastController(BaseController):
end=True,
naive=True,
export_pred_raw=True,
metric=True,
)
ns_parser = self.parse_known_args_and_warn(
parser,
@@ -2415,6 +2434,7 @@ class ForecastController(BaseController):
end_date=ns_parser.s_end_date,
naive=ns_parser.naive,
export_pred_raw=ns_parser.export_pred_raw,
metric=ns_parser.metric,
)
@log_start_end(log=logger)
@@ -2482,6 +2502,7 @@ class ForecastController(BaseController):
end=True,
naive=True,
export_pred_raw=True,
metric=True,
)
ns_parser = self.parse_known_args_and_warn(
parser,
@@ -2525,6 +2546,7 @@ class ForecastController(BaseController):
end_date=ns_parser.s_end_date,
naive=ns_parser.naive,
export_pred_raw=ns_parser.export_pred_raw,
metric=ns_parser.metric,
)
@log_start_end(log=logger)
@@ -2561,6 +2583,7 @@ class ForecastController(BaseController):
naive=True,
explainability_raw=True,
export_pred_raw=True,
metric=True,
)
ns_parser = self.parse_known_args_and_warn(
parser,
@@ -2595,6 +2618,7 @@ class ForecastController(BaseController):
naive=ns_parser.naive,
explainability_raw=ns_parser.explainability_raw,
export_pred_raw=ns_parser.export_pred_raw,
metric=ns_parser.metric,
)
@log_start_end(log=logger)
@@ -2630,6 +2654,7 @@ class ForecastController(BaseController):
naive=True,
explainability_raw=True,
export_pred_raw=True,
metric=True,
)
ns_parser = self.parse_known_args_and_warn(
parser,
@@ -2663,6 +2688,7 @@ class ForecastController(BaseController):
naive=ns_parser.naive,
explainability_raw=ns_parser.explainability_raw,
export_pred_raw=ns_parser.export_pred_raw,
metric=ns_parser.metric,
)
@log_start_end(log=logger)
@@ -2714,6 +2740,7 @@ class ForecastController(BaseController):
end=True,
naive=True,
export_pred_raw=True,
metric=True,
)
ns_parser = self.parse_known_args_and_warn(
parser,
@@ -2755,6 +2782,7 @@ class ForecastController(BaseController):
end_date=ns_parser.s_end_date,
naive=ns_parser.naive,
export_pred_raw=ns_parser.export_pred_raw,
metric=ns_parser.metric,
)
@log_start_end(log=logger)
@@ -2842,6 +2870,7 @@ class ForecastController(BaseController):
end=True,
naive=True,
export_pred_raw=True,
metric=True,
)
ns_parser = self.parse_known_args_and_warn(
parser,
@@ -2885,6 +2914,7 @@ class ForecastController(BaseController):
end_date=ns_parser.s_end_date,
naive=ns_parser.naive,
export_pred_raw=ns_parser.export_pred_raw,
metric=ns_parser.metric,
)
@log_start_end(log=logger)
@@ -2958,6 +2988,7 @@ class ForecastController(BaseController):
end=True,
naive=True,
export_pred_raw=True,
metric=True,
)
ns_parser = self.parse_known_args_and_warn(
parser,
@@ -3001,6 +3032,7 @@ class ForecastController(BaseController):
end_date=ns_parser.s_end_date,
naive=ns_parser.naive,
export_pred_raw=ns_parser.export_pred_raw,
metric=ns_parser.metric,
)
@log_start_end(log=logger)
@@ -3092,6 +3124,7 @@ class ForecastController(BaseController):
end=True,
naive=True,
export_pred_raw=True,
metric=True,
)
ns_parser = self.parse_known_args_and_warn(
parser,
@@ -3136,6 +3169,7 @@ class ForecastController(BaseController):
end_date=ns_parser.s_end_date,
naive=ns_parser.naive,
export_pred_raw=ns_parser.export_pred_raw,
metric=ns_parser.metric,
)
@log_start_end(log=logger)

32
openbb_terminal/forecast/helpers.py
View File

@@ -11,7 +11,7 @@ import pandas as pd
from darts import TimeSeries
from darts.dataprocessing.transformers import MissingValuesFiller, Scaler
from darts.explainability.shap_explainer import ShapExplainer
from darts.metrics import mape
from darts.metrics import mape, mse, rmse, smape
from darts.models.forecasting.torch_forecasting_model import GlobalForecastingModel
from darts.utils.statistics import plot_residuals_analysis
from pytorch_lightning.callbacks.early_stopping import EarlyStopping
@@ -477,6 +477,7 @@ def plot_forecast(
forecast_only: bool = False,
naive: bool = False,
export_pred_raw: bool = False,
metric: str = "mape",
external_axes: Optional[List[plt.axes]] = None,
):
quant_kwargs = {}
@@ -503,7 +504,16 @@ def plot_forecast(
if naive:
# show naive forecast shift timeseries by 1
naive_fcast = naive_fcast.drop_before(historical_fcast.start_time())
naive_precision = mape(ticker_series, naive_fcast)
# calculate precision based on metric
if metric == "rsme":
naive_precision = rmse(ticker_series, naive_fcast)
elif metric == "mse":
naive_precision = mse(ticker_series, naive_fcast)
elif metric == "mape":
naive_precision = mape(ticker_series, naive_fcast)
elif metric == "smape":
naive_precision = smape(ticker_series, naive_fcast)
naive_fcast.plot(
label=f"Naive+1: {naive_precision:.2f}%",
@@ -516,7 +526,7 @@ def plot_forecast(
pred_label += " w/ past covs"
predicted_values.plot(label=pred_label, **quant_kwargs, color="#00AAFF")
ax.set_title(
f"{name} for <{ticker_name}> for next [{n_predict}] days (MAPE={precision:.2f}%)"
f"{name} for <{ticker_name}> for next [{n_predict}] days ({metric.upper()}={precision:.2f}%)"
)
ax.set_ylabel(target_col)
ax.set_xlabel("Date")
@@ -719,6 +729,7 @@ def get_prediction(
train_split: float,
forecast_horizon: int,
n_predict: int,
metric: str,
):
_, val = ticker_series.split_before(train_split)
@@ -771,10 +782,17 @@ def get_prediction(
else:
prediction = best_model.predict(series=ticker_series, n=n_predict)
precision = mape(
actual_series=val, pred_series=historical_fcast
) # mape = mean average percentage error
console.print(f"{model_name} model obtains MAPE: {precision:.2f}% \n")
# calculate precision based on metric (rmse, mse, mape)
if metric == "rmse":
precision = rmse(actual_series=val, pred_series=historical_fcast)
elif metric == "mse":
precision = mse(actual_series=val, pred_series=historical_fcast)
elif metric == "mape":
precision = mape(actual_series=val, pred_series=historical_fcast)
elif metric == "smape":
precision = smape(actual_series=val, pred_series=historical_fcast)
console.print(f"{model_name} model obtains {metric.upper()}: {precision:.2f}% \n")
# scale back
if use_scalers and isinstance(scaler, Scaler):

4
openbb_terminal/forecast/linregr_model.py
View File

@@ -27,6 +27,7 @@ def get_linear_regression_data(
output_chunk_length: int = 5,
lags: Union[int, List[int]] = 14,
random_state: Optional[int] = None,
metric: str = "mape",
) -> Tuple[
List[TimeSeries],
List[TimeSeries],
@@ -56,6 +57,8 @@ def get_linear_regression_data(
lagged target values to predict the next time step
random_state: Optional[int]
The state for the model
metric: str
The metric to use for the model. Defaults to "mape".
Returns
-------
@@ -112,4 +115,5 @@ def get_linear_regression_data(
train_split,
forecast_horizon,
n_predict,
metric,
)

9
openbb_terminal/forecast/linregr_view.py
View File

@@ -35,6 +35,7 @@ def display_linear_regression(
naive: bool = False,
explainability_raw: bool = False,
export_pred_raw: bool = False,
metric: str = "mape",
external_axes: Optional[List[plt.axes]] = None,
):
"""Display Linear Regression Forecasting
@@ -74,6 +75,12 @@ def display_linear_regression(
naive: bool
Whether to show the naive baseline. This just assumes the closing price will be the
same as the previous day's closing price. Defaults to False.
explainability_raw: bool
Whether to show the raw explainability data. Defaults to False.
export_pred_raw: bool
Whether to export the raw prediction data. Defaults to False.
metric: str
The metric to use for the model. Defaults to "mape".
external_axes: Optional[List[plt.axes]]
External axes to plot on
"""
@@ -100,6 +107,7 @@ def display_linear_regression(
forecast_horizon=forecast_horizon,
output_chunk_length=output_chunk_length,
lags=lags,
metric=metric,
)
probabilistic = True
@@ -121,6 +129,7 @@ def display_linear_regression(
forecast_only=forecast_only,
naive=naive,
export_pred_raw=export_pred_raw,
metric=metric,
external_axes=external_axes,
)
if residuals:

4
openbb_terminal/forecast/nbeats_model.py
View File

@@ -37,6 +37,7 @@ def get_NBEATS_data(
model_save_name: str = "nbeats_model",
force_reset: bool = True,
save_checkpoints: bool = True,
metric: str = "mape",
) -> Tuple[
Optional[List[TimeSeries]],
Optional[List[TimeSeries]],
@@ -88,6 +89,8 @@ def get_NBEATS_data(
save_checkpoints: bool
Whether or not to automatically save the untrained model and checkpoints from training.
Defaults to True.
metric: str
Metric to use for model selection. Defaults to "mape".
Returns
-------
@@ -168,4 +171,5 @@ def get_NBEATS_data(
train_split,
forecast_horizon,
n_predict,
metric,
)

7
openbb_terminal/forecast/nbeats_view.py
View File

@@ -44,6 +44,7 @@ def display_nbeats_forecast(
end_date: Optional[datetime] = None,
naive: bool = False,
export_pred_raw: bool = False,
metric: str = "mape",
external_axes: Optional[List[plt.axes]] = None,
):
"""Display NBEATS forecast
@@ -108,6 +109,10 @@ def display_nbeats_forecast(
naive: bool
Whether to show the naive baseline. This just assumes the closing price will be the same
as the previous day's closing price. Defaults to False.
export_pred_raw: bool
Whether to export the raw predicted values. Defaults to False.
metric: str
Metric to use for evaluation. Defaults to "mape".
external_axes: Optional[List[plt.axes]]
External axes to plot on
"""
@@ -145,6 +150,7 @@ def display_nbeats_forecast(
model_save_name=model_save_name,
force_reset=force_reset,
save_checkpoints=save_checkpoints,
metric=metric,
)
if ticker_series == []:
return
@@ -168,6 +174,7 @@ def display_nbeats_forecast(
forecast_only=forecast_only,
naive=naive,
export_pred_raw=export_pred_raw,
metric=metric,
external_axes=external_axes,
)
if residuals:

4
openbb_terminal/forecast/nhits_model.py
View File

@@ -43,6 +43,7 @@ def get_nhits_data(
model_save_name: str = "nhits_model",
force_reset: bool = True,
save_checkpoints: bool = True,
metric: str = "mape",
) -> Tuple[
Optional[List[TimeSeries]],
Optional[List[TimeSeries]],
@@ -109,6 +110,8 @@ def get_nhits_data(
discarded). Defaults to True.
save_checkpoints: bool
Whether or not to automatically save the untrained model and checkpoints from training. Defaults to True.
metric: str
Metric to use for model selection. Defaults to "mape".
Returns
-------
@@ -197,4 +200,5 @@ def get_nhits_data(
train_split,
forecast_horizon,
n_predict,
metric,
)

7
openbb_terminal/forecast/nhits_view.py
View File

@@ -49,6 +49,7 @@ def display_nhits_forecast(
end_date: Optional[datetime] = None,
naive: bool = False,
export_pred_raw: bool = False,
metric: str = "mape",
external_axes: Optional[List[plt.axes]] = None,
):
"""Display Nhits forecast
@@ -128,6 +129,10 @@ def display_nhits_forecast(
naive: bool
Whether to show the naive baseline. This just assumes the closing price will be the same
as the previous day's closing price. Defaults to False.
export_pred_raw: bool
Whether to export the raw predicted values. Defaults to False.
metric: str
Metric to use for evaluation. Defaults to "mape".
external_axes: Optional[List[plt.axes]]
External axes to plot on
"""
@@ -169,6 +174,7 @@ def display_nhits_forecast(
model_save_name=model_save_name,
force_reset=force_reset,
save_checkpoints=save_checkpoints,
metric=metric,
)
if ticker_series == []:
return
@@ -192,6 +198,7 @@ def display_nhits_forecast(
forecast_only=forecast_only,
naive=naive,
export_pred_raw=export_pred_raw,
metric=metric,
external_axes=external_axes,
)
if residuals:

4
openbb_terminal/forecast/regr_model.py
View File

@@ -26,6 +26,7 @@ def get_regression_data(
forecast_horizon: int = 5,
output_chunk_length: int = 5,
lags: Union[int, List[int]] = 14,
metric: str = "mape",
) -> Tuple[
List[TimeSeries], List[TimeSeries], List[TimeSeries], float, type[RegressionModel]
]:
@@ -49,6 +50,8 @@ def get_regression_data(
The length of the forecast of the model. Defaults to 1.
lags: Union[int, List[int]]
lagged target values to predict the next time step
metric: str
Metric to use for evaluation. Defaults to "mape".
Returns
-------
@@ -99,4 +102,5 @@ def get_regression_data(
train_split,
forecast_horizon,
n_predict,
metric,
)

5
openbb_terminal/forecast/regr_view.py
View File

@@ -35,6 +35,7 @@ def display_regression(
naive: bool = False,
explainability_raw: bool = False,
export_pred_raw: bool = False,
metric: str = "mape",
external_axes: Optional[List[plt.axes]] = None,
):
"""Display Regression Forecasting
@@ -74,6 +75,8 @@ def display_regression(
naive: bool
Whether to show the naive baseline. This just assumes the closing price will be the same
as the previous day's closing price. Defaults to False.
metric: str
The metric to use for the forecast. Defaults to "mape".
external_axes: Optional[List[plt.axes]]
External axes to plot on
"""
@@ -100,6 +103,7 @@ def display_regression(
forecast_horizon=forecast_horizon,
output_chunk_length=output_chunk_length,
lags=lags,
metric=metric,
)
probabilistic = False
helpers.plot_forecast(
@@ -120,6 +124,7 @@ def display_regression(
forecast_only=forecast_only,
naive=naive,
export_pred_raw=export_pred_raw,
metric=metric,
external_axes=external_axes,
)
if residuals:

4
openbb_terminal/forecast/rnn_model.py
View File

@@ -36,6 +36,7 @@ def get_rnn_data(
input_chunk_size: int = 14,
force_reset: bool = True,
save_checkpoints: bool = True,
metric: str = "mape",
) -> Tuple[
Optional[List[type[TimeSeries]]],
Optional[List[type[TimeSeries]]],
@@ -77,6 +78,8 @@ def get_rnn_data(
save_checkpoints: bool
Whether or not to automatically save the untrained model and checkpoints from training.
Defaults to True.
metric: str
Metric to use for model selection. Defaults to "mape".
Returns
-------
@@ -145,4 +148,5 @@ def get_rnn_data(
train_split,
forecast_horizon,
n_predict,
metric,
)

7
openbb_terminal/forecast/rnn_view.py
View File

@@ -42,6 +42,7 @@ def display_rnn_forecast(
end_date: Optional[datetime] = None,
naive: bool = False,
export_pred_raw: bool = False,
metric: str = "mape",
external_axes: Optional[List[plt.axes]] = None,
):
"""Display RNN forecast
@@ -94,6 +95,10 @@ def display_rnn_forecast(
naive: bool
Whether to show the naive baseline. This just assumes the closing price will be the same
as the previous day's closing price. Defaults to False.
export_pred_raw: bool
Whether to export the raw predictions. Defaults to False.
metric: str
The metric to use for the forecast. Defaults to "mape".
external_axes: Optional[List[plt.axes]]
External axes to plot on
"""
@@ -124,6 +129,7 @@ def display_rnn_forecast(
input_chunk_size=input_chunk_size,
force_reset=force_reset,
save_checkpoints=save_checkpoints,
metric=metric,
)
if ticker_series == []:
return
@@ -148,6 +154,7 @@ def display_rnn_forecast(
forecast_only=forecast_only,
naive=naive,
export_pred_raw=export_pred_raw,
metric=metric,
external_axes=external_axes,
)
if residuals:

4
openbb_terminal/forecast/tcn_model.py
View File

@@ -37,6 +37,7 @@ def get_tcn_data(
model_save_name: str = "tcn_model",
force_reset: bool = True,
save_checkpoints: bool = True,
metric: str = "mape",
) -> Tuple[
List[Optional[TimeSeries]],
List[Optional[TimeSeries]],
@@ -86,6 +87,8 @@ def get_tcn_data(
save_checkpoints: bool
Whether or not to automatically save the untrained model and checkpoints from training.
Defaults to True.
metric: str
Metric to use for model selection. Defaults to "mape".
Returns
-------
@@ -166,4 +169,5 @@ def get_tcn_data(
train_split,
forecast_horizon,
n_predict,
metric,
)

7
openbb_terminal/forecast/tcn_view.py
View File

@@ -44,6 +44,7 @@ def display_tcn_forecast(
end_date: Optional[datetime] = None,
naive: bool = False,
export_pred_raw: bool = False,
metric: str = "mape",
external_axes: Optional[List[plt.axes]] = None,
):
"""Display TCN forecast
@@ -104,6 +105,10 @@ def display_tcn_forecast(
naive: bool
Whether to show the naive baseline. This just assumes the closing price will be the same
as the previous day's closing price. Defaults to False.
export_pred_raw: bool
Whether to export the raw predicted values. Defaults to False.
metric: str
The metric to use for the model. Defaults to "mape".
external_axes: Optional[List[plt.axes]]
External axes to plot on
"""
@@ -141,6 +146,7 @@ def display_tcn_forecast(
model_save_name=model_save_name,
force_reset=force_reset,
save_checkpoints=save_checkpoints,
metric=metric,
)
if ticker_series == []:
return
@@ -165,6 +171,7 @@ def display_tcn_forecast(
naive=naive,
export_pred_raw=export_pred_raw,
external_axes=external_axes,
metric=metric,
)
if residuals:
helpers.plot_residuals(

4
openbb_terminal/forecast/tft_model.py
View File

@@ -43,6 +43,7 @@ def get_tft_data(
model_save_name: str = "tft_model",
force_reset: bool = True,
save_checkpoints: bool = True,
metric: str = "mape",
) -> Tuple[
Optional[List[TimeSeries]],
Optional[List[TimeSeries]],
@@ -101,6 +102,8 @@ def get_tft_data(
save_checkpoints: (bool, optional)
Whether or not to automatically save the untrained model and checkpoints from training.
Defaults to True.
metric: (str, optional)
Metric to use for model selection. Defaults to "mape".
Returns
-------
@@ -204,4 +207,5 @@ def get_tft_data(
train_split,
forecast_horizon,
n_predict,
metric,
)

5
openbb_terminal/forecast/tft_view.py
View File

@@ -45,6 +45,7 @@ def display_tft_forecast(
end_date: Optional[datetime] = None,
naive: bool = False,
export_pred_raw: bool = False,
metric: str = "mape",
external_axes: Optional[List[plt.axes]] = None,
):
"""Display Temporal Fusion Transformer forecast
@@ -105,6 +106,8 @@ def display_tft_forecast(
naive: bool
Whether to show the naive baseline. This just assumes the closing price will be the same
as the previous day's closing price. Defaults to False.
metric: str
The metric to use for the model. Defaults to "mape".
external_axes:Optional[List[plt.axes]]
External axes to plot on
"""
@@ -143,6 +146,7 @@ def display_tft_forecast(
model_save_name=model_save_name,
force_reset=force_reset,
save_checkpoints=save_checkpoints,
metric=metric,
)
if ticker_series == []:
return
@@ -166,6 +170,7 @@ def display_tft_forecast(
forecast_only=forecast_only,
naive=naive,
export_pred_raw=export_pred_raw,
metric=metric,
external_axes=external_axes,
)
if residuals:

28
openbb_terminal/forecast/theta_model.py
View File

@@ -8,7 +8,7 @@ from typing import List, Optional, Tuple, Union
import numpy as np
import pandas as pd
from darts import TimeSeries
from darts.metrics import mape
from darts.metrics import mape, mse, rmse, smape
from darts.models import Theta
from darts.utils.utils import SeasonalityMode
from statsmodels.tools.sm_exceptions import ConvergenceWarning
@@ -34,6 +34,7 @@ def get_theta_data(
n_predict: int = 5,
start_window: float = 0.85,
forecast_horizon: int = 5,
metric: str = "mape",
) -> Tuple[
Optional[List[TimeSeries]],
Optional[List[TimeSeries]],
@@ -64,7 +65,9 @@ def get_theta_data(
start_window: float
Size of sliding window from start of timeseries and onwards
forecast_horizon: int
Number of days to forecast when backtesting and retraining historical
Number of days to forecast when backtesting and retraining historical data
metric: str
Metric to use when backtesting and retraining historical data. Defaults to "mape".
Returns
-------
@@ -135,10 +138,23 @@ def get_theta_data(
# fit model on entire series for final prediction
best_theta_model_final.fit(ticker_series)
prediction = best_theta_model_final.predict(int(n_predict))
precision = mape(
actual_series=ticker_series, pred_series=historical_fcast_theta
) # mape = mean average precision error
console.print(f"Theta Model obtains MAPE: {precision:.2f}% \n")
if metric == "rmse":
precision = rmse(
actual_series=ticker_series, pred_series=historical_fcast_theta
)
elif metric == "mse":
precision = mse(actual_series=ticker_series, pred_series=historical_fcast_theta)
elif metric == "mape":
precision = mape(
actual_series=ticker_series, pred_series=historical_fcast_theta
)
elif metric == "smape":
precision = smape(
actual_series=ticker_series, pred_series=historical_fcast_theta
)
console.print(f"Theta Model obtains {metric.upper()}: {precision:.2f}% \n")
return (
ticker_series,

7
openbb_terminal/forecast/theta_view.py
View File

@@ -34,6 +34,7 @@ def display_theta_forecast(
end_date: Optional[datetime] = None,
naive: bool = False,
export_pred_raw: bool = False,
metric: str = "mape",
external_axes: Optional[List[plt.axes]] = None,
):
"""Display Theta forecast
@@ -73,6 +74,10 @@ def display_theta_forecast(
naive: bool
Whether to show the naive baseline. This just assumes the closing price will be the same
as the previous day's closing price. Defaults to False.
export_pred_raw: bool
Whether to export the raw predicted values. Defaults to False.
metric: str
The metric to use for backtesting. Defaults to "mape".
external_axes:Optional[List[plt.axes]]
External axes to plot on
"""
@@ -95,6 +100,7 @@ def display_theta_forecast(
target_column=target_column,
start_window=start_window,
forecast_horizon=forecast_horizon,
metric=metric,
)
if ticker_series == []:
return
@@ -119,6 +125,7 @@ def display_theta_forecast(
naive=naive,
export_pred_raw=export_pred_raw,
external_axes=external_axes,
metric=metric,
)
if residuals:

4
openbb_terminal/forecast/trans_model.py
View File

@@ -40,6 +40,7 @@ def get_trans_data(
model_save_name: str = "trans_model",
force_reset: bool = True,
save_checkpoints: bool = True,
metric: str = "mape",
) -> Tuple[
Optional[List[TimeSeries]],
Optional[List[TimeSeries]],
@@ -94,6 +95,8 @@ def get_trans_data(
discarded). Defaults to True.
save_checkpoints: bool
Whether or not to automatically save the untrained model and checkpoints from training. Defaults to True.
metric: str
Metric to use for model selection. Defaults to "mape".
Returns
-------
@@ -176,4 +179,5 @@ def get_trans_data(
train_split,
forecast_horizon,
n_predict,
metric,
)

7
openbb_terminal/forecast/trans_view.py
View File

@@ -47,6 +47,7 @@ def display_trans_forecast(
end_date: Optional[datetime] = None,
naive: bool = False,
export_pred_raw: bool = False,
metric: str = "mape",
external_axes: Optional[List[plt.axes]] = None,
):
"""Display Transformer forecast
@@ -112,6 +113,10 @@ def display_trans_forecast(
naive: bool
Whether to show the naive baseline. This just assumes the closing price will be the same
as the previous day's closing price. Defaults to False.
export_pred_raw: bool
Whether to export the raw predicted values. Defaults to False.
metric: str
The metric to use for the model. Defaults to "mape".
external_axes: Optional[List[plt.axes]]
External axes to plot on
"""
@@ -152,6 +157,7 @@ def display_trans_forecast(
model_save_name=model_save_name,
force_reset=force_reset,
save_checkpoints=save_checkpoints,
metric=metric,
)
if ticker_series == []:
return
@@ -176,6 +182,7 @@ def display_trans_forecast(
naive=naive,
export_pred_raw=export_pred_raw,
external_axes=external_axes,
metric=metric,
)
if residuals:
helpers.plot_residuals(