forked from huawei/mindspore2022
modify doc
This commit is contained in:
parent
c670b2f472
commit
28b876f20d
|
|
@ -108,7 +108,7 @@ Boost能够自动加速网络,如减少BN/梯度冻结/累积梯度等。
|
|||
|
||||
其中:
|
||||
|
||||
- pca_mat (array): 维度(k*n),k是n_components的大小,n是权重的大小。
|
||||
- pca_mat (array): 维度(k*n),k是 *n_components* 的大小,n是权重的大小。
|
||||
- bk (array): 维度(k*k),bk是拟牛顿法中的对称正定矩阵。
|
||||
|
||||
我们需要找到满足以下条件的m:
|
||||
|
|
@ -116,7 +116,7 @@ Boost能够自动加速网络,如减少BN/梯度冻结/累积梯度等。
|
|||
.. math::
|
||||
new\_loss < old\_loss + delta\_loss
|
||||
|
||||
然后使用delta_grad去更新模型的权重:
|
||||
然后使用 *delta_grad* 去更新模型的权重:
|
||||
|
||||
.. math::
|
||||
|
||||
|
|
|
|||
|
|
@ -75,7 +75,7 @@ class TFRecordToMR:
|
|||
source (str): TFRecord file to be transformed.
|
||||
destination (str): MindRecord file path to transform into, ensure that no file with the same name
|
||||
exists in the directory.
|
||||
feature_dict (dict[str, `FixedLenFeature <https://www.tensorflow.org/api_docs/python/tf/io/FixedLenFeature>`_]): Dictionary
|
||||
feature_dict (dict[str, `FixedLenFeature <https://www.tensorflow.org/api_docs/python/tf/io/FixedLenFeature>`_ ]): Dictionary
|
||||
that states the feature type, and `VarLenFeature <https://www.tensorflow.org/api_docs/python/tf/io/VarLenFeature>`_
|
||||
is not supported.
|
||||
bytes_fields (list[str], optional): The bytes fields which are in `feature_dict` and can be images bytes.
|
||||
|
|
|
|||
Loading…
Reference in New Issue