-
Hi, when i use the command "python -m torch.distributed.run --nproc_per_node=8 pretrain.py --config ./configs/Pretrain.yaml --output_dir output/Pretrain ".
It shows "ERROR:torch.distributed.elastic.m…
-
Hi, I would like to ask about incorporating additional training objectives that are beneficial to downstream tasks during the pre-training of BGE on top of the MLM task.
Specifically, my downstream…
-
Hi, I saw in the article that you reported the results of pre-training on SSV2, but I could not find its pre-training script and checkpoint. Could you please provide it? Or did I not find the link? Lo…
wjj-w updated
4 months ago
-
Thank you for the excellent work! But I'm having difficulty reproducing the results on DHF1k using diff-sal.
I’ve downloaded the pre-trained checkpoint on DHF1k provided in this repository, but I’m…
-
Do you have trained model weights? I would greatly appreciate it if you could provide it!
-
Hello! I am very inspired by your work. Referring to your work, I have some doubts while pre-training MRI data.
I want to use brain tumor MRI containing four modalities for pre-training, what to do …
-
Hi,
I have a question:
Will Bleu be improved if the pre-trained word vectors are embedded in the neural machine translation model for retraining?
Looking forward to your advice or answers.
Best…
-
Wenet2.2.1 version, the test set is librispeech's test_clean, the test results, the official pre-training model u2++conformer, two questions:
1. recognize.py uses attention mode to decode, and a larg…
-
Why is the loss still huge when loading pre-training weights for training, just like not loading them?
-
From the paper also, in the 4.2 ablation study, I don't find any clear element why the network need to be trained in two stages (DMTimg and DMTvid) especially cause you have claimed:
> Therefore, w…