add categorical class #27695
add categorical class #27695
Conversation
|
Thanks for your contribution! |
| * :math:`[x=i]` : it evaluates to 1 if :math:`x==i` , 0 otherwise. | ||
|
|
||
| Args: | ||
| logits(list|numpy.ndarray|Tensor): The logits input of categorical distribution. The data type is float32 or float64. |
There was a problem hiding this comment.
add parameter name, will refine it in next PR.
There was a problem hiding this comment.
done, refer to PR #27946
| def __init__(self, logits, name=None): | ||
| """ | ||
| Args: | ||
| logits(list|numpy.ndarray|Variable): The logits input of categorical distribution. The data type is float32 or float64. |
There was a problem hiding this comment.
Variable change to Tensor, add parameter name, will refine it in next PR.
There was a problem hiding this comment.
done, refer to PR #27946
| other (Categorical): instance of Categorical. The data type is float32. | ||
|
|
||
| Returns: | ||
| Variable: kl-divergence between two Categorical distributions. |
There was a problem hiding this comment.
change Variable to Tensor, will refine it in next PR.
There was a problem hiding this comment.
done, refer to PR #27946
| """Shannon entropy in nats. | ||
|
|
||
| Returns: | ||
| Variable: Shannon entropy of Categorical distribution. The data type is float32. |
There was a problem hiding this comment.
change Variable to Tensor, will refine it in next PR.
There was a problem hiding this comment.
done, refer to PR #27946
| Tensor: A tensor with prepended dimensions shape. | ||
|
|
||
| Examples: | ||
| .. code-block:: python |
There was a problem hiding this comment.
``code-block:: python` needs indentation. The same below.
There was a problem hiding this comment.
done, refer to PR #27946
PR types
Others
PR changes
APIs
Describe
add
CategoricalclassIt's a special case of
multinomial( refer to PR add multinomial op #27219 ).Categoricaldistribution is a discrete probability distribution that describes the possible results of a random variable that can take on one of K possible categories, with the probability of each category separately specified.add
samplemethodUsing
multinomialto sample fromCategoricaldistribution.add
probsandlog_probmethodGiven category index, return the probability of this category.