U sVc@sdZddlmZddlmZddlZddlmZddl m Z m Z erPddl m Z dd d d d d ZdddddddZd d dddd dddZdS)zH Module containing utilities for NDFrame.sample() and .GroupBy.sample() ) annotations) TYPE_CHECKINGN)lib) ABCDataFrame ABCSeries)NDFramerintz np.ndarray)objaxisreturnc Cst|tr||j|}t|trt|tr||dkrrz ||}Wqztk rn}ztd|W5d}~XYqzXqtdntdt|tr|j}n|j }||ddj }t ||j |krtdt |rtd |dkrtd t|}|r|}d||<|S) z Process and validate the `weights` argument to `NDFrame.sample` and `.GroupBy.sample`. Returns `weights` as an ndarray[np.float64], validated except for normalizing weights (because that must be done groupwise in groupby sampling). rz+String passed to weights not a valid columnNzLStrings can only be passed to weights when sampling from rows on a DataFramez@Strings cannot be passed as weights when sampling from a Series.Zfloat64)Zdtypez5Weights and axis to be sampled must be of same lengthz*weight vector may not include `inf` valuesz.weight vector many not include negative values) isinstancerZreindexZaxesstrrKeyError ValueErrorZ _constructorZ_constructor_slicedZ_valueslenshaperZhas_infsanynpisnancopy)r weightsr errfuncmissingr6/tmp/pip-unpacked-wheel-xj8nt62q/pandas/core/sample.pypreprocess_weightssD         rz int | Nonez float | Nonebool)nfracreplacer cCs|dkr|dkrd}nx|dk r0|dk r0tdn^|dk r^|dkrHtd|ddkrtdn0|dk sjt|dkr~|s~td|dkrtd|S) z Process and validate the `n` and `frac` arguments to `NDFrame.sample` and `.GroupBy.sample`. Returns None if `frac` should be used (variable sampling sizes), otherwise returns the constant sampling size. Nz0Please enter a value for `frac` OR `n`, not bothrz=A negative number of rows requested. Please provide `n` >= 0.z$Only integers accepted as `n` valueszJReplace has to be set to `True` when upsampling the population `frac` > 1.z@A negative number of rows requested. Please provide `frac` >= 0.)rAssertionError)rrr rrrprocess_sampling_sizeNs*      r#znp.ndarray | Nonez+np.random.RandomState | np.random.Generator)obj_lensizer r random_stater cCsH|dk r*|}|dkr"||}ntd|j||||djtjddS)ac Randomly sample `size` indices in `np.arange(obj_len)` Parameters ---------- obj_len : int The length of the indices being considered size : int The number of values to choose replace : bool Allow or disallow sampling of the same row more than once. weights : np.ndarray[np.float64] or None If None, equal probability weighting, otherwise weights according to the vector normalized random_state: np.random.RandomState or np.random.Generator State used for the random sampling Returns ------- np.ndarray[np.intp] Nrz$Invalid weights: weights sum to zero)r%r pF)r)sumrchoiceZastyperZintp)r$r%r rr&Z weight_sumrrrsampless r*)__doc__ __future__rtypingrZnumpyrZ pandas._libsrZpandas.core.dtypes.genericrrZpandas.core.genericrrr#r*rrrrs    9%