swapnil-lader
swapnil-lader
It worked thanks
> Could you provide a bit more details how the dataloader and export function was called? Thanks. import torch import model_navigator as nav import argparse import gradio as gr import...
> We pushed new version which contains commands logging. If you could re-run the export and share log that would help us investigate the issues. Thanks! 2022-09-27 15:02:35,649 INFO Navigator...
> Could you also paste the info logs from before the error? Especially for the Infer metadata output command. Also, what is the output structure of your model? Is it...
I am adding a little more details about the decoder layer and output info if this helps. (decoder): BARTDecoder( (model): MBartForCausalLM( (model): MBartDecoderWrapper( (decoder): MBartDecoder( (embed_tokens): Embedding(57530, 1024, padding_idx=1) (embed_positions):...
> I am adding a little more details about the decoder layer and output info if this helps. (decoder): BARTDecoder( (model): MBartForCausalLM( (model): MBartDecoderWrapper( (decoder): MBartDecoder( (embed_tokens): Embedding(57530, 1024, padding_idx=1)...
@jkosek and @ptarasiewiczNV So I tried multiple different methods, but I cannot create the correct data loader for two different model layers combined as one in the DONUT model. I...
> @caleb-recursion Thank you for this example. I was able to reproduce the error. Here are the conclusions. > > The way that `fasterrcnn_mobilenet_v3_large_320_fpn` handles batching is not currently supported...