79 lines
3.1 KiB
Python
79 lines
3.1 KiB
Python
from typing import Any, Dict
|
|
import pandas as pd
|
|
import numpy as np
|
|
import torch
|
|
|
|
class PandasFeatureSplitToPt:
|
|
"""
|
|
Splits a Pandas DataFrame into feature PyTorch tensor and label PyTorch tensor.
|
|
If output_1d_label is set, label will be output as a 1d array instead of a 2d array.
|
|
|
|
category: Transformation
|
|
"""
|
|
|
|
@classmethod
|
|
def INPUT_TYPES(cls) -> Dict[str, Any]:
|
|
"""
|
|
Defines the input types for the function.
|
|
|
|
Returns:
|
|
Dict[str, Any]: A dictionary specifying required input types.
|
|
"""
|
|
return {
|
|
"required": {
|
|
"dataframe": ("DATAFRAME", {}),
|
|
"label_integer_position": ("INT", {"default": -1, "min": -10, "max": 1024}),
|
|
"output_1d_label": ("BOOLEAN", {"default": True}),
|
|
"feature_data_type": (("float32","float16", "bloat16", "float64", "int8", "uint8", "int16", "int32", "int64"),),
|
|
"label_data_type": (("float32","float16", "bloat16", "float64", "int8", "uint8", "int16", "int32", "int64"),)
|
|
}
|
|
}
|
|
|
|
RETURN_TYPES: tuple = ("TENSOR", "TENSOR")
|
|
RETURN_NAMES: tuple = ("feature", "label")
|
|
|
|
FUNCTION: str = "f"
|
|
CATEGORY: str = "Data Analysis"
|
|
|
|
def f(self, dataframe: pd.DataFrame,
|
|
label_integer_position: int,
|
|
output_1d_label: bool,
|
|
feature_data_type: str,
|
|
label_data_type: str) -> tuple:
|
|
"""
|
|
Splits a Pandas DataFrame into feature Numpy ndarray and label Numpy ndarray.
|
|
|
|
Args:
|
|
dataframe (pd.DataFrame): The DataFrame to split.
|
|
label_integer_position (int): The column integer position of the label.
|
|
output_1d_label (bool): If output_1d_label is set, label will be output as a 1d array instead of a 2d array.
|
|
feature_data_type (str): Feature data type.
|
|
label_data_type (str): Label data type.
|
|
Returns:
|
|
tuple: A tuple containing the feature and label arrays.
|
|
"""
|
|
feature_dtype = getattr(torch, feature_data_type)
|
|
label_dtype = getattr(torch, label_data_type)
|
|
|
|
num_columns = dataframe.shape[1]
|
|
|
|
if label_integer_position < 0:
|
|
label_integer_position += num_columns
|
|
|
|
# Validate label_integer_position
|
|
if label_integer_position >= num_columns:
|
|
raise ValueError(f"label_integer_position {label_integer_position} is out of range for dataframe with {num_columns} columns.")
|
|
|
|
# Extract features
|
|
np_left = dataframe.iloc[:, :label_integer_position].to_numpy()
|
|
np_right = dataframe.iloc[:, label_integer_position + 1:].to_numpy() if label_integer_position < num_columns - 1 else np.empty((len(dataframe), 0))
|
|
np_features = np.hstack([np_left, np_right]) if np_right.shape[1] > 0 else np_left
|
|
|
|
# Extract label
|
|
np_label = dataframe.iloc[:, label_integer_position].to_numpy()
|
|
if output_1d_label:
|
|
np_label = np_label.ravel()
|
|
|
|
return (torch.tensor(np_features,dtype=feature_dtype),
|
|
torch.tensor(np_label, dtype=label_dtype))
|