about interpolation

[Winnie202]

I trained my own dataset, but the effect of interpolation is not good. If I do not want to interpolation frames, that is, input frame=4,output frame=4, how should I modify it? can i modify the output_frame in the weight of completed training?

[llmpass]

You may also need to modify the positional encoding part, otherwise it will not work.

[Winnie202]

You may also need to modify the positional encoding part, otherwise it will not work.

Excuse me, which parts of the code are modified

[llmpass]

You may also need to modify the positional encoding part, otherwise it will not work.

Excuse me, which parts of the code are modified

https://github.com/llmpass/RSTT/blob/b10988958137cc9fa8a2b814f0f94c4900d736fc/models/layers.py#L163

[Winnie202]

You may also need to modify the positional encoding part, otherwise it will not work.

Excuse me, which parts of the code are modified

https://github.com/llmpass/RSTT/blob/b10988958137cc9fa8a2b814f0f94c4900d736fc/models/layers.py#L163

Excuse me, don't i need to modify the output sequence length at the following code section? Because num_out_frames is defined here,I'm confused about this,because I don't need to change the size of the output feature map, I just need to modify it num_out_frames same as num_input_frames

[llmpass]

You may also need to modify the positional encoding part, otherwise it will not work.

Excuse me, which parts of the code are modified

https://github.com/llmpass/RSTT/blob/b10988958137cc9fa8a2b814f0f94c4900d736fc/models/layers.py#L163

Excuse me, don't i need to modify the output sequence length at the following code section? Because num_out_frames is defined here,I'm confused about this,because I don't need to change the size of the output feature map, I just need to modify it num_out_frames same as num_input_frames

yes.

[Winnie202]

You may also need to modify the positional encoding part, otherwise it will not work.

Excuse me, which parts of the code are modified

https://github.com/llmpass/RSTT/blob/b10988958137cc9fa8a2b814f0f94c4900d736fc/models/layers.py#L163

Excuse me, don't i need to modify the output sequence length at the following code section? Because num_out_frames is defined here,I'm confused about this,because I don't need to change the size of the output feature map, I just need to modify it num_out_frames same as num_input_frames

yes.

if I don't need to change the size of the output feature map, I just need to modify it num_out_frames same as num_input_frames,does the following code section need change

[llmpass]

You may also need to modify the positional encoding part, otherwise it will not work.

Excuse me, which parts of the code are modified

https://github.com/llmpass/RSTT/blob/b10988958137cc9fa8a2b814f0f94c4900d736fc/models/layers.py#L163

Excuse me, don't i need to modify the output sequence length at the following code section? Because num_out_frames is defined here,I'm confused about this,because I don't need to change the size of the output feature map, I just need to modify it num_out_frames same as num_input_frames

yes.

if I don't need to change the size of the output feature map, I just need to modify it num_out_frames same as num_input_frames,does the following code section need change

You are right.