search

maxjcohen / transformer

Implementation of Transformer model (originally from Attention is All You Need) applied to Time Series.
https://timeseriestransformer.readthedocs.io/en/latest/
GNU General Public License v3.0
852 stars 166 forks source link

A question #58

Closed chuzheng88 closed 2 years ago

chuzheng88 commented 2 years ago

image

why did the error occur ?

maxjcohen commented 2 years ago

Hi, please take a minute to read the documentation. The shape of the input tensor should be (batch_size, sequence_length, input_dimension), so in your case (batch_size, 1000, 2).

chuzheng88 commented 2 years ago

image

also error occur

maxjcohen commented 2 years ago

Again, please read the documentation of MultiHeadAttentionChunk. You're using the Chunk MHA, which clearly states that the default chunk_size is 168. This is why you are getting an error.