photo

Zhiyu WANG


Last seen: 2ヶ月 前 |  2022 年からアクティブ

Followers: 0   Following: 0

Programming Languages:
C
Spoken Languages:
English

統計

  • Thankful Level 1
  • First Review
  • Knowledgeable Level 1
  • First Answer
  • Explorer

バッジを表示

Feeds

表示方法

質問


is it possible to make "Sequence-to-Sequence Classification" using custom CNN network?
If so , how to create a sequence datastore and what`s the dataformat of the corresponding dlarray object?

2年以上 前 | 1 件の回答 | 0

1

回答

回答済み
SequenceinputLayer and Convolution2dLayer
It`s recommended to transform your data to formatted dlarray objects at first. As for the dlarray object, you can refer to this...

2年以上 前 | 0

回答済み
using predictAndUpdateState with an LSTM with featureInputLayer
LSTM requires somelike "timeseries" data while "featureInputLayer" is not along with that.Yo ucan refer to this answer: https:/...

2年以上 前 | 0

| 採用済み

回答済み
Does the input size of sequenceInputLayer have to be set 1?
Becasue the input sequence data is a 1 by 10 double.It menas each time you take 1 token out of the sequence.

2年以上 前 | 0