-
Hi,
I am trying to fine-tune blip2 with coco. In the caption_dataset there is a sample['input_text']. However the forward method for blip2_t5 expects sample['output_text']. The prompt is being attach…
-
Hi,
I was trying the "zero copy" method in the t5 notebook on a seq2seq transformer model. When I set the `clone_tensor` to True everything looks fine, just not as much speedup as I expected.
W…
-
如题,持续关注中
-
Hi, I've been trying to deploy an mtf model to the NVIDIA Triton Inference Server by converting the SavedModel (output of model.export()) to an onnx file with no luck. I've been receiving several erro…
-
Hi thanks for this great library! I'm trying to get BLIP-2 running on Replicate, using `blip2_t5` with `caption_coco_flant5xl`. I was able to get the original BLIP working fine, but when I try BLIP-2 …
-
Provide as much information as possible. At least, this should include a description of your issue and steps to reproduce the problem. If possible also provide a summary of what steps or workarounds y…
-
Thanks for sharing the repo . It is really helpful.
I'm exploring ways to do the optimization on GPU. I know its not presently supported. Could you share some approach or references to implement th…
-
### System Info
x85-64
4 A10
0.9.0
### Who can help?
_No response_
### Information
- [X] The official example scripts
- [ ] My own modified scripts
### Tasks
- [ ] An officially supported tas…
-
Hi Philipp,
thanks for your awesome blog on training Flan T5 XXL. I am playing around with it and doing just zero-shot inference using ds_flan_t5_z3_config_bf16.json deepspeed config file. I believ…
-
### Feature request
Hi, for `AutoTokenizer.train_new_from_iterator` there's a hardcoded `tqdm` progress bar I want to swap to `rich` and I'm happy to PR it back.
I can see on github it's at `trans…