Huaying Yuan

There is a line in file preprocessing/convert_formats/msmarco_doc_create_train_input_orig.py: `from matchmaker.evaluation.msmarco_eval import *` but there is no evaluation filefolder in matchmaker, could you be kind enough to release the code for this...

Does FastSeq support video generation models such as Latte?

3

Hi, thank you for making this open source. I've noticed that parameters such as 'sequence_parallel_size' and 'sequence_parallel_group' only appear in 'DiT' modules (such as 'DistAttn') but not in 'Latte' modules....

Questions about longseq

3

Hi, thanks for your great work. I've conducted tests on two A40 machines utilizing accelerators. Executing the following command: ``` torchrun --standalone --nproc_per_node=2 train.py \ --model DiT-XL/2 \ --batch_size 3...

225x90x90

1

Hello, I noticed you mentioned that the latest code could support training with a latent size of 225x90x90, which seems quite large. However, I couldn't find the corresponding training script...

Hello, I am trying to preprocess text in your way: https://github.com/Victorwz/VaLM/blob/161accd153943a502ec7e3983c448dda07f9d975/myscripts/preprocess_valm_text.sh#L4C1-L5C1 However, there is no file called 'clip.vocab' in "./data". Could you tell me how to get it?

Where to get the pretrained model with max-seq-length over 512?

4

I am trying to train a ltp model tackling long document, but where can I get the pretrained model with max-seq-length over 512? As far as I know, pretrained models...

vqa_models misjudge the spatial relationship in most of the cases

1

Hello, thank you for your repo. When I test some cases with tifa, I find that the model totally confused right and left, below is an example: what I generate...

Hello, thank you for your open-source work!! I wanted to ask about the part mentioned in your paper where the OFA model was used to generate frame-level captions. Were these...