salesforce/LAVIS
LAVIS - A One-stop Library for Language-Vision Intelligence
Jupyter NotebookBSD-3-Clause
Issues
- 0
FileNotFoundError: Package has no location <module 'imageio_ffmpeg.binaries' (namespace)>
#709 opened by davidhouse2023 - 8
BLIP-2 paper finetune replicate low performance: BLEU_4 score is 0.15 for finetuning COCO_caption
#707 opened by LuoyaoChen - 0
KeyError: 'image_id'
#708 opened by mshmoon - 1
A question for BLIP2 Q-Former
#706 opened by nbqu - 1
- 2
generated output error
#704 opened by Mei0211 - 3
How should I use blip2 for vqa task training?
#688 opened by WildLight - 0
How can I change cache_root?
#703 opened by minnsu03 - 0
cache_version Value Error
#702 opened by minnsu03 - 0
Blip2-caption only generate "a photo of"
#697 opened by VincentWangty - 0
How is `Total Params` calculated?
#701 opened by Wuzimeng - 2
safetensors_rust.SafetensorError: Error while deserializing header: HeaderTooLarge
#693 opened by slrhs - 0
CUstom dataset Inference
#700 opened by aliman80 - 0
- 0
About Text Preprocessing of InstructBLIP
#698 opened by davidwang200099 - 3
- 1
- 1
Why do I always encounter CUDA out of memory problem when I load the load_model_process function? Can the RTX 3090 be used for the BLIP-2 model?"
#670 opened by zhangmenghuan-mh - 1
Use BLIP-2 for Image Captioning
#692 opened by ArefAz - 0
- 0
salesforce-lavis 1.0.2 requires transformers<4.27,>=4.25.0, but you have transformers 4.40.0 which is incompatible.
#691 opened by CS123n - 3
- 0
XInstructBLIP demo text generation
#689 opened by ParkJun-Yeong - 1
how to deal with “Missing keys ”
#686 opened by jackbrown333 - 0
huggingface_hub.utils._validators.HFValidationError: Repo id must use alphanumeric chars or '-', '_', '.', '--' and '..' are forbidden, '-' and '.' cannot start or end the name, max length is 96: ''.
#687 opened by WildLight - 2
- 0
how use it output target class。
#685 opened by skydh - 0
How can I calculate the similarity between multimodal features and Unimodal features
#681 opened by dsdsknfsk - 0
Input of multiple images
#684 opened by dsdsknfsk - 0
- 5
OPT2.7B underperforming & weird behavior compared to flant5xl on image captioning?
#676 opened by Thomas2419 - 0
How to run InstructBLIP with other LLM model
#680 opened by zhangzitenga - 0
Image use to present LAVIS
#679 opened by Raphaelle-Lemaire - 0
The role of modeling_opt.py in the BLIP2 model
#677 opened by abinzzz - 0
The results of DocVQA, infoVQA, and OCRVQA evaluating the instructblip model are very low
#671 opened by Fym68 - 0
Score difference in ITM and ITC ?
#672 opened by Kapil-23 - 0
- 1
instruct-blip output long meanless string
#666 opened by leedewdew - 0
question about text localization
#667 opened by Yorkev - 0
- 1
Error occurred during BLIP2-demo execution
#663 opened by Kim-DKyu - 0
When using instructblip to evaluate the okvqa data set, there is nothing in the output path
#664 opened by Fym68 - 1
How does VQA works on BLIP2 without LLM?
#659 opened by jihwanp - 5
- 0
Why checkpoints are not saved?
#662 opened by SnowNation101 - 0
About the lables in the VQA task using OPT models
#661 opened by dszpr - 0
Segmentation Fault
#660 opened by dfan - 0
How to extract features from custom finetuned BLIP2 model on HuggingFace?
#657 opened by SnowNation101 - 0
The relevant code for this job of img2llm
#656 opened by wangfengjuan - 0