mindspore 1.10.0__cp37-none-any.whl → 2.0.0rc1__cp37-none-any.whl

mindspore/nn/learning_rate_schedule.py

@@ -21,7 +21,7 @@ import math
 from mindspore.common import dtype as mstype
 from mindspore.ops import operations as P
 from mindspore.nn.cell import Cell
-from mindspore._checkparam import Validator as validator
+from mindspore import _checkparam as validator
 
 
 class LearningRateSchedule(Cell):
@@ -133,7 +133,7 @@ class NaturalExpDecayLR(LearningRateSchedule):
     For current step, the formula of computing decayed learning rate is:
 
     .. math::
-        decayed\_learning\_rate= learning\_rate * e^{-decay\_rate * p}
+        decayed\_learning\_rate = learning\_rate * e^{-decay\_rate * p}
 
     Where :
 
@@ -273,9 +273,8 @@ class CosineDecayLR(LearningRateSchedule):
     For current step, the formula of computing decayed learning rate is:
 
     .. math::
-        decayed\_learning\_rate = min\_lr + 0.5 * (max\_lr - min\_lr) *
-        (1 + cos(\frac{current\_step}{decay\_steps}\pi))
-
+        decayed\_learning\_rate = &min\_lr + 0.5 * (max\_lr - min\_lr) *\\
+        &(1 + cos(\frac{current\_step}{decay\_steps}\pi))
 
     Args:
         min_lr (float): The minimum value of learning rate.
@@ -343,8 +342,9 @@ class PolynomialDecayLR(LearningRateSchedule):
     For current step, the formula of computing decayed learning rate is:
 
     .. math::
-        decayed\_learning\_rate = (learning\_rate - end\_learning\_rate) *
-        (1 - tmp\_step / tmp\_decay\_steps)^{power} + end\_learning\_rate
+        decayed\_learning\_rate = &(learning\_rate - end\_learning\_rate) *\\
+        &(1 - tmp\_step / tmp\_decay\_steps)^{power}\\
+        &+ end\_learning\_rate
 
     Where :
 
@@ -461,7 +461,7 @@ class WarmUpLR(LearningRateSchedule):
         ValueError: If `learning_rate` is less than or equal to 0.
 
     Supported Platforms:
-        ``Ascend`` ``GPU``
+        ``Ascend`` ``GPU`` ``CPU``
 
     Examples:
         >>> import mindspore

mindspore/nn/loss/__init__.py

@@ -20,13 +20,16 @@ It shows how well the model works on a dataset and the optimization target which
 """
 from __future__ import absolute_import
 
-from mindspore.nn.loss.loss import LossBase, L1Loss, MSELoss, SmoothL1Loss, SoftMarginLoss, FocalLoss,\
+from mindspore.nn.loss.loss import LossBase, L1Loss, CTCLoss, MSELoss, SmoothL1Loss, SoftMarginLoss, FocalLoss, \
     SoftmaxCrossEntropyWithLogits, BCELoss, MultiMarginLoss, CosineEmbeddingLoss, \
-    SampledSoftmaxLoss, DiceLoss, BCEWithLogitsLoss, MultiClassDiceLoss,\
-    RMSELoss, MAELoss, HuberLoss, CrossEntropyLoss, NLLLoss
+    SampledSoftmaxLoss, PoissonNLLLoss, MultiLabelSoftMarginLoss, DiceLoss, BCEWithLogitsLoss, MultiClassDiceLoss, \
+    RMSELoss, MAELoss, HuberLoss, CrossEntropyLoss, NLLLoss, KLDivLoss, MarginRankingLoss, GaussianNLLLoss, \
+    HingeEmbeddingLoss, MultilabelMarginLoss, TripletMarginLoss
 
 
-__all__ = ['LossBase', 'L1Loss', 'MSELoss', 'SmoothL1Loss', 'SoftMarginLoss', 'FocalLoss',
+__all__ = ['LossBase', 'L1Loss', 'CTCLoss', 'MSELoss', 'SmoothL1Loss', 'SoftMarginLoss', 'FocalLoss',
            'SoftmaxCrossEntropyWithLogits', 'BCELoss', 'BCEWithLogitsLoss', 'MultiMarginLoss',
-           'CosineEmbeddingLoss', 'SampledSoftmaxLoss', 'DiceLoss', 'MultiClassDiceLoss',
-           'RMSELoss', 'MAELoss', 'HuberLoss', 'CrossEntropyLoss', 'NLLLoss']
+           'CosineEmbeddingLoss', 'SampledSoftmaxLoss', 'PoissonNLLLoss',
+           'MultiLabelSoftMarginLoss', 'DiceLoss', 'MultiClassDiceLoss', 'MultilabelMarginLoss',
+           'RMSELoss', 'MAELoss', 'HuberLoss', 'CrossEntropyLoss', 'NLLLoss', 'KLDivLoss', 'MarginRankingLoss',
+           'GaussianNLLLoss', 'HingeEmbeddingLoss', 'TripletMarginLoss']