refac: fix uuid

src/wavy/panel.py

@@ -2,6 +2,7 @@
 
 # import contextlib
 import random
+import uuid
 import warnings
 from itertools import chain
 
@@ -50,27 +51,32 @@ def create_panels(
 
  ids = np.arange(lookback, end)
 
- xframes = np.zeros(shape=(len(ids) * lookback, df.shape[1]))
+ xframes = np.empty(shape=(len(ids) * lookback, df.shape[1]), dtype="object")
  xindex = [
- np.zeros(shape=(len(ids) * lookback), dtype=int),
- np.zeros(shape=(len(ids) * lookback), dtype=df.index.dtype),
+ np.empty(shape=(len(ids) * lookback), dtype="<U31"),
+ np.empty(shape=(len(ids) * lookback), dtype=df.index.dtype),
  ]
 
- yframes = np.zeros(shape=(len(ids) * horizon, df.shape[1]))
+ yframes = np.empty(shape=(len(ids) * horizon, df.shape[1]), dtype="object")
  yindex = [
- np.zeros(shape=(len(ids) * horizon), dtype=int),
- np.zeros(shape=(len(ids) * horizon), dtype=df.index.dtype),
+ np.empty(shape=(len(ids) * horizon), dtype="<U31"),
+ np.empty(shape=(len(ids) * horizon), dtype=df.index.dtype),
  ]
 
  for i in ids:
+ uuid_str = str(uuid.uuid4())
+
  # X
  frame = df.iloc[i - lookback : i]
  xframes[
  (i - lookback) * lookback : (i - lookback + 1) * lookback, :
- ] = frame.values # BUG: Fail when df contains strings.
- xindex[0][(i - lookback) * lookback : (i - lookback + 1) * lookback] = (
- i - lookback
- ) * np.ones(lookback, dtype=int)
+ ] = frame.values
+ # xindex[0][(i - lookback) * lookback : (i - lookback + 1) * lookback] = (
+ # i - lookback
+ # ) * np.ones(lookback, dtype=int)
+ xindex[0][
+ (i - lookback) * lookback : (i - lookback + 1) * lookback
+ ] = np.repeat(uuid_str, lookback)
  xindex[1][
  (i - lookback) * lookback : (i - lookback + 1) * lookback
  ] = frame.index.values
@@ -80,9 +86,12 @@ def create_panels(
  yframes[
  (i - lookback) * horizon : (i - lookback + 1) * horizon, :
  ] = frame.values
- yindex[0][(i - lookback) * horizon : (i - lookback + 1) * horizon] = (
- i - lookback
- ) * np.ones(horizon, dtype=int)
+ # yindex[0][(i - lookback) * horizon : (i - lookback + 1) * horizon] = (
+ # i - lookback
+ # ) * np.ones(horizon, dtype=int)
+ yindex[0][(i - lookback) * horizon : (i - lookback + 1) * horizon] = np.repeat(
+ uuid_str, horizon
+ )
  yindex[1][
  (i - lookback) * horizon : (i - lookback + 1) * horizon
  ] = frame.index.values
@@ -100,27 +109,26 @@ def create_panels(
  )
 
 
-def reset_ids(panels: list[Panel], inplace: bool = False) -> list[Panel]:
- """
- Reset ids of a panel.
-
- Args:
- panels (``list``): List of panels
- inplace (``bool``): Whether to reset ids inplace or not.
+# def reset_ids(panels: list[Panel], inplace: bool = False) -> list[Panel]:
+# """
+# Reset ids of a panel.
 
- Returns:
- ``Panel``: Reset id of panel
- """
+#  Args:
+#  panels (``list``): List of panels
+#  inplace (``bool``): Whether to reset ids inplace or not.
 
- # Check if id in x and y are the same
- # ! Broken
- # if not all(np.array_equal(panels[0].ids, panel.ids) for panel in panels):
- # raise ValueError(
- # "Ids for panels are not the same. Try using match function first."
- # )
+# Returns:
+# ``Panel``: Reset id of panel
+# """
 
+# # Check if id in x and y are the same
+# # ! Broken
+# # if not all(np.array_equal(panels[0].ids, panel.ids) for panel in panels):
+# # raise ValueError(
+# # "Ids for panels are not the same. Try using match function first."
+# # )
 
- return [panel.reset_ids(inplace=inplace) for panel in panels]
+#  return [panel.reset_ids(inplace=inplace) for panel in panels]
 
 
 def dropna_match(x, y):
@@ -200,15 +208,20 @@ def set_training_split(
  >>> x, y = set_training_split(x, y, train_size=0.8, val_size=0.2, test_size=0.1)
  """
 
- x.set_training_split(train_size=train_size, val_size=val_size, test_size=test_size)
-
- # ! set_train_split varies for different num_timesteps, thus x and y cannot be created using it. Removed the line below:
- # y.set_training_split(train_size=train_size, val_size=val_size, test_size=test_size)
-
- # ! Added instead:
- y.train_size = x.train_size
- y.val_size = x.val_size
- y.test_size = x.test_size
+ if x.num_timesteps >= y.num_timesteps:
+ x.set_training_split(
+ train_size=train_size, val_size=val_size, test_size=test_size
+ )
+ y.train_size = x.train_size
+ y.val_size = x.val_size
+ y.test_size = x.test_size
+ else:
+ y.set_training_split(
+ train_size=train_size, val_size=val_size, test_size=test_size
+ )
+ x.train_size = y.train_size
+ x.val_size = y.val_size
+ x.test_size = y.test_size
 
 
 class _PanelSeries(pd.Series):
@@ -536,7 +549,6 @@ def set_training_split(
  self.val_size = n_val - self.num_timesteps + 1
  self.test_size = n_test - self.num_timesteps + 1
 
-
  def to_dataframe(self) -> pd.DataFrame:
  """
  Convert panel to dataframe.
@@ -830,35 +842,35 @@ def shuffle_panel(self, seed: int = None, reset_ids: bool = False) -> Panel | No
  return new_panel
 
  # ! Inconsistent with lookback >= 2 if frames were modified.
- def plot(
- self,
- add_annotation: bool = True,
- max: int = 10_000,
- use_timestep: bool = False,
- **kwargs,
- ) -> plot.PanelFigure:
- """
- Plot the panel.
-
- Args:
- add_annotation (``bool``): If True, plot the training, validation, and test annotation.
- max (``int``): Maximum number of samples to plot.
- use_timestep (``bool``): If True, plot the timestep instead of the sample index.
- **kwargs: Additional arguments to pass to the plot function.
-
- Returns:
- ``plot``: Result of plot function.
- """
-
- panel = self.row_panel(n=0)
-
- if max and self.num_frames > max:
- return plot(
- panel.sample_panel(max, how="spaced"),
- use_timestep=use_timestep,
- add_annotation=add_annotation,
- **kwargs,
- )
- return plot(
- panel, use_timestep=use_timestep, add_annotation=add_annotation, **kwargs
- )
+ # def plot(
+ #  self,
+ #  add_annotation: bool = True,
+ #  max: int = 10_000,
+ #  use_timestep: bool = False,
+ #  **kwargs,
+ # ) -> plot.PanelFigure:
+ #  """
+ #  Plot the panel.
+
+ #  Args:
+ #  add_annotation (``bool``): If True, plot the training, validation, and test annotation.
+ #  max (``int``): Maximum number of samples to plot.
+ #  use_timestep (``bool``): If True, plot the timestep instead of the sample index.
+ #  **kwargs: Additional arguments to pass to the plot function.
+
+ #  Returns:
+ #  ``plot``: Result of plot function.
+ #  """
+
+ #  panel = self.row_panel(n=0)
+
+ #  if max and self.num_frames > max:
+ #  return plot(
+ #  panel.sample_panel(max, how="spaced"),
+ #  use_timestep=use_timestep,
+ #  add_annotation=add_annotation,
+ #  **kwargs,
+ #  )
+ #  return plot(
+ #  panel, use_timestep=use_timestep, add_annotation=add_annotation, **kwargs
+ #  )