Referencia de la Clase vsm.model.ldacgsmulti.LdaCgsMulti

Diagrama de herencias de vsm.model.ldacgsmulti.LdaCgsMulti

Diagrama de colaboración para vsm.model.ldacgsmulti.LdaCgsMulti:

Métodos públicos
def	__init__
def	word_top
def	word_top
def	inv_top_sums
def	inv_top_sums
def	top_doc
def	top_doc
def	corpus
def	corpus
def	Z
def	Z
def	K
def	K
def	V
def	V
def	iteration
def	iteration
def	train
Métodos públicos heredados desde vsm.model.ldacgsseq.LdaCgsSeq
def	__init__
def	Z_split
def	docs
def	train
def	save

Métodos públicos estáticos
def	load
Métodos públicos estáticos heredados desde vsm.model.ldacgsseq.LdaCgsSeq
def	load

Atributos públicos
	V
	corpus
	Z
	word_top
	inv_top_sums
	top_doc
	iteration
Atributos públicos heredados desde vsm.model.ldacgsseq.LdaCgsSeq
	context_type
	K
	V
	indices
	corpus
	Z
	alpha
	word_top
	inv_top_sums
	top_doc
	iteration
	log_probs
	seed

Atributos públicos estáticos
tuple	maxint = np.iinfo(np.int32)
list	seeds = [np.random.randint(0, maxint) for n in range(n_proc)]
	K = beta)
tuple	pbar = ProgressBar(widgets=[Percentage(), Bar()], maxval=n_iterations)
int	iteration = 0
tuple	data = zip(docs, doc_indices, self._mtrand_states)
tuple	results = p.map(update, data)
tuple	lp = np.sum(logp_ls)

Descripción detallada

An implementation of LDA using collapsed Gibbs sampling with multi-processing.

On Windows platforms, LdaCgsMulti is not supported. A NotImplementedError 
will be raised notifying the user to use the LdaCgsSeq package. Users
desiring a platform-independent fallback should use LDA(multiprocess=True) to
initialize the object, which will return either a LdaCgsMulti or a LdaCgsSeq
instance, depending on the platform, while raising a RuntimeWarning.

Documentación del constructor y destructor

def vsm.model.ldacgsmulti.LdaCgsMulti.__init__	(		self,
			corpus = None,
			context_type = None,
			K = 20,
			V = 0,
			Z = [],
			alpha = [],
			beta = [],
			n_proc = 2,
			seeds = None
	)

Initialize LdaCgsMulti.

:param corpus: Source of observed data.
:type corpus: `Corpus`
    
:param context_type: Name of tokenization stored in `corpus` whose tokens
    will be treated as documents.
:type context_type: string, optional

:param K: Number of topics. Default is `20`.
:type K: int, optional
    
:param alpha: Context priors. Default is a flat prior of 0.01 
    for all contexts.
:type alpha: list, optional
    
:param beta: Topic priors. Default is 0.01 for all topics.
:type beta: list, optional

:param n_proc: Number of processors used for training. Default is
    2.
:type n_proc: int, optional

:param seeds: List of random seeds, one for each thread.
    The length of the list should be same as `n_proc`. Default is `None`.
:type seeds: list of integers, optional

Documentación de las funciones miembro

def vsm.model.ldacgsmulti.LdaCgsMulti.load (   filename )

static

A static method for loading a saved LdaCgsMulti model.

:param filename: Name of a saved model to be loaded.
:type filename: string

:returns: m : LdaCgsMulti object

:See Also: :class:`numpy.load`

def vsm.model.ldacgsmulti.LdaCgsMulti.train	(		self,
			n_iterations = 500,
			verbose = 1
	)

Takes an optional argument, `n_iterations` and updates the model
`n_iterations` times.

:param n_iterations: Number of iterations. Default is 500.
:type n_iterations: int, optional

:param verbose: If `True`, current number of iterations
    are printed out to notify the user. Default is `True`.
:type verbose: boolean, optional

Documentación de los datos miembro

tuple vsm.model.ldacgsmulti.LdaCgsMulti.maxint = np.iinfo(np.int32)

static

LdaCgsMulti is not implemented on 
Windows. Please use LdaCgsSeq.

---

La documentación para esta clase fue generada a partir del siguiente fichero:

• vsm/vsm/model/ldacgsmulti.py