salesforce/BLIP issues and pull requests

#226 - captioner filter

Issue - State: open - Opened by cheun726 15 days ago

#225 - ITG Loss Adjustment

Issue - State: open - Opened by QVQZZZ about 1 month ago

#224 - Access Denied for answer list for the test dataset for vqa dataset

Issue - State: open - Opened by amitpadegal about 2 months ago - 1 comment

#223 - run demo on colab bug

Issue - State: open - Opened by zkailinzhang about 2 months ago - 1 comment

#222 - Repetitive Output for All images

Issue - State: closed - Opened by bruceMug 2 months ago

#221 - What key design processes make BLIP have no limitations on text input length compared to CLIP?

Issue - State: open - Opened by Yang-bug-star 3 months ago

#220 - How can I do inference with pretrained models rather than finetuned models

Issue - State: open - Opened by Yang-bug-star 3 months ago

#219 - What does the model(**inputs, output_attentions=True) output?

Issue - State: open - Opened by ZhanliangAaronWang 4 months ago

#218 - FileNotFoundError: [Errno 2] No such file or directory: 'export/share/datasets/vision/coco/images/val2014/COCO_val2014_000000184613.jpg'

Issue - State: open - Opened by Jingut 4 months ago - 1 comment

#217 - How to train itm from itc

Issue - State: open - Opened by Raion-Shin 5 months ago

#216 - is it possible to change the text LLM of blip to a different LLM?

Issue - State: open - Opened by KAABSHAHID 5 months ago - 1 comment

#215 - I have created a multimodal large model technology exchange group,welcome to join us.

Issue - State: open - Opened by feihuamantian 5 months ago

#214 - Caption on ImageNet-Dogs

Issue - State: closed - Opened by LouisDong95 5 months ago

#213 - fix the wrong words

Pull Request - State: open - Opened by ArtificialZeng 6 months ago - 2 comments
Labels: cla:signed

#212 - knowledge distillation

Issue - State: open - Opened by sssssshf 6 months ago

#211 - blip_vqa error

Issue - State: open - Opened by AIWASS23 6 months ago - 2 comments

#210 - fixed runtime device error

Pull Request - State: open - Opened by antelopoul 8 months ago
Labels: cla:signed

#209 - How to use roberta as the decoder

Issue - State: open - Opened by xiweideng 9 months ago

#208 - The rank_answer function in BLIP is different from that in ALBEF

Issue - State: open - Opened by littleFlyDance 9 months ago

#207 - Question or bug in blip_pretrain.py

Issue - State: open - Opened by LiGuo12 9 months ago

#206 - How to retrive the raw attention scores or logits from blip model ( image captioning)

Issue - State: open - Opened by umme17 9 months ago

#205 - I want to use the existing image-text pedestrian dataset and finetune the BLIP model. Should I use pre-trained checkpoints weights or finetuned checkpoints weights?

Issue - State: open - Opened by shams2023 10 months ago

#205 - I want to use the existing image-text pedestrian dataset and finetune the BLIP model. Should I use pre-trained checkpoints weights or finetuned checkpoints weights?

Issue - State: open - Opened by shams2023 10 months ago

#204 - Image-Text Retrieval

Issue - State: open - Opened by mjjc111 10 months ago

#204 - Image-Text Retrieval

Issue - State: open - Opened by mjjc111 10 months ago

#203 - LAION 115M dataset has 11164.tar?

Issue - State: open - Opened by jacob-kang 10 months ago

#203 - LAION 115M dataset has 11164.tar?

Issue - State: open - Opened by jacob-kang 10 months ago

#202 - Error while running Colab demo

Issue - State: open - Opened by staru09 11 months ago - 1 comment

#201 - stable-diffusion RuntimeError: Couldn't fetch BLIP.

Issue - State: open - Opened by saiheitor 11 months ago - 1 comment

#201 - stable-diffusion RuntimeError: Couldn't fetch BLIP.

Issue - State: open - Opened by saiheitor 11 months ago - 1 comment

#200 - ITM Loss Stuck at 0.63

Issue - State: open - Opened by bfan1256 11 months ago - 3 comments

#200 - ITM Loss Stuck at 0.63

Issue - State: open - Opened by bfan1256 11 months ago - 2 comments

#199 - How to use the retrival large model for image-text prediction (model_large_retrieval_coco) ?

Issue - State: open - Opened by caydenwei 12 months ago

#199 - How to use the retrival large model for image-text prediction (model_large_retrieval_coco) ?

Issue - State: open - Opened by caydenwei 12 months ago

#198 - Blip Replicate Interface Is Down

Issue - State: closed - Opened by hashnimo 12 months ago - 1 comment

#198 - Blip Replicate Interface Is Down

Issue - State: closed - Opened by hashnimo 12 months ago - 1 comment

#197 - ModuleNotFoundError: No module named 'ruamel_yaml'

Issue - State: open - Opened by WenjunHuang94 12 months ago - 2 comments

#196 - web demo issue

Issue - State: open - Opened by hhzhao0525 12 months ago

#196 - web demo issue

Issue - State: open - Opened by hhzhao0525 12 months ago

#195 - BlipForImageTextRetrieval loss returned when labels is provided

Issue - State: open - Opened by AIIRLab 12 months ago

#195 - BlipForImageTextRetrieval loss returned when labels is provided

Issue - State: open - Opened by AIIRLab 12 months ago

#194 - RuntimeError: Default process group has not been initialized, please make sure to call init_process_group (train_caption.py)

Issue - State: open - Opened by Y-HuiMing-Y about 1 year ago - 1 comment

#193 - what's the effect of 'image_queue' and 'text_queue'

Issue - State: open - Opened by FengWu-PKU about 1 year ago - 1 comment

#193 - what's the effect of 'image_queue' and 'text_queue'

Issue - State: open - Opened by FengWu-PKU about 1 year ago - 1 comment

#192 - How does the BLIP model use half-precision (FP16) inference?

Issue - State: open - Opened by WKaiH123 about 1 year ago

#192 - How does the BLIP model use half-precision (FP16) inference?

Issue - State: open - Opened by WKaiH123 about 1 year ago

#191 - About the ViT of BLIP

Issue - State: open - Opened by LWShowTime about 1 year ago

#191 - About the ViT of BLIP

Issue - State: open - Opened by LWShowTime about 1 year ago

#190 - Need clearly Understand of each checkpoint

Issue - State: open - Opened by p1k0pan about 1 year ago

#190 - Need clearly Understand of each checkpoint

Issue - State: open - Opened by p1k0pan about 1 year ago

#189 - I am having trouble running evaluation code

Issue - State: closed - Opened by jyrana about 1 year ago - 5 comments

#188 - 相似图像生成的caption一样，该如何解决？

Issue - State: open - Opened by shams2023 about 1 year ago

#187 - Video subtitle generation

Issue - State: open - Opened by 497329284 about 1 year ago

#186 - 一张3090卡去微调COCO检索，需要多长时间？

Issue - State: open - Opened by shams2023 over 1 year ago - 2 comments

#185 - Add auto-label guide

Pull Request - State: closed - Opened by capjamesg over 1 year ago - 2 comments
Labels: cla:missing

#184 - New ViT findings via registers (2309.16588)

Issue - State: open - Opened by Infinitay over 1 year ago

#183 - The pre-trained BLIP model is used directly to perform caption operation, but the generated caption effect is not good

Issue - State: closed - Opened by shams2023 over 1 year ago

#182 - .

Issue - State: closed - Opened by shams2023 over 1 year ago

#181 - No scores of VQA evaluation

Issue - State: closed - Opened by p1k0pan over 1 year ago - 5 comments

#180 - This error indicates that your module has parameters that were not used in producing loss

Issue - State: open - Opened by ericosmic over 1 year ago

#179 - Request for BLIP Pretrain Training Logs and Dataset Inquiry

Issue - State: open - Opened by Aitical over 1 year ago

#178 - what is mean of 'question_states += [question_output.last_hidden_state[b]]*n'

Issue - State: open - Opened by ericosmic over 1 year ago

#177 - retrieve output not fix

Issue - State: open - Opened by ltm920716 over 1 year ago

#176 - ERROR: Could not build wheels for tokenizers, which is required to install pyproject.toml-based projects

Issue - State: open - Opened by TheOneTrueGuy over 1 year ago - 4 comments

#175 - Can BLIP generate more words image caption?

Issue - State: open - Opened by uestcMeng over 1 year ago - 7 comments

#174 - Cosine between image_features and text_features taken from BLIP_Extractor_Features gives bad results

Issue - State: open - Opened by aTunass over 1 year ago

#173 - demo.ipynb : RuntimeError: The size of tensor a (3) must match the size of tensor b (9) at non-singleton dimension 0

Issue - State: closed - Opened by Taiga10969 over 1 year ago - 6 comments

#172 - Questions when evaluating the finetuned BLIP model on COCO.

Issue - State: open - Opened by dhhxhhh over 1 year ago

#171 - Update med.py: Fixed the issue of BERTEncoder.forward() not returning cross-attentions when requested

Pull Request - State: open - Opened by programmingLearner over 1 year ago - 1 comment
Labels: cla:missing

#170 - I train on Chinese data with 5000w image-text pairs and it works.

Issue - State: open - Opened by Hoogck over 1 year ago

#169 - Convert BLIP model to TensorRT

Issue - State: open - Opened by Frostbite22 over 1 year ago - 1 comment

#168 - RuntimeError: Input type (torch.cuda.FloatTensor) and weight type (torch.FloatTensor) should be the same

Issue - State: open - Opened by HWH-2000 over 1 year ago

#167 - Image-Text Matching result werid

Issue - State: closed - Opened by jucic over 1 year ago

#166 - Issue with Tracing PyTorch Model Using torch.jit.trace

Issue - State: open - Opened by delegatepattern over 1 year ago

#165 - The size of tensor a (3) must match the size of tensor b (9) at non-singleton dimension 0

Issue - State: open - Opened by Peter-D-James over 1 year ago - 19 comments

#164 - Pre-training with LAION

Issue - State: open - Opened by aries-young over 1 year ago

#163 - Batch size for pre-training

Issue - State: open - Opened by aries-young over 1 year ago

#162 - Bump transformers from 4.15.0 to 4.30.0

Pull Request - State: open - Opened by dependabot[bot] over 1 year ago
Labels: dependencies

#161 - RuntimeError: The size of tensor a (6) must match the size of tensor b (18) at non-singleton dimension 0

Issue - State: closed - Opened by gpx333 over 1 year ago - 1 comment

#160 - Update requirements.txt error by installing transformers version

Pull Request - State: open - Opened by cobanov over 1 year ago - 4 comments
Labels: cla:missing

#159 - AMP dtype incompatibility

Issue - State: closed - Opened by danieltudosiu over 1 year ago

#158 - Error 'RuntimeError: indices should be either on cpu or on the same device as the indexed tensor (cpu) ' in line 147 of train_retrieval.py

Issue - State: open - Opened by Qusijia over 1 year ago - 1 comment

#157 - Ruamel import is outdated

Issue - State: open - Opened by derekmil over 1 year ago

#156 - Extracting attention value

Issue - State: open - Opened by NamHyelin over 1 year ago

#155 - BLIP Image Captioning GradCAM?

Issue - State: open - Opened by gwyong over 1 year ago - 8 comments

#154 - How did you deal with some extremely duplicated captions in the pretraining dataset?

Issue - State: open - Opened by SCZwangxiao over 1 year ago

#153 - Can this model output Chinese image captionings?

Issue - State: open - Opened by K-tang-mkv over 1 year ago - 3 comments

#152 - The ITM and LM loss do not converge

Issue - State: open - Opened by chenyzh28 over 1 year ago - 3 comments

#151 - Demo Google Colab not working

Issue - State: open - Opened by detrin over 1 year ago - 6 comments

#150 - Reproduce image captioning results

Issue - State: open - Opened by yxoh over 1 year ago

#149 - How to load the pre-trained BLIP model pth file into HuggingFace BLIP model?

Issue - State: open - Opened by adventure2165 over 1 year ago - 1 comment

#148 - Validation time more than the training time

Issue - State: open - Opened by aman-cc over 1 year ago

#147 - About zero-shot Image-Text Retrieval

Issue - State: open - Opened by rookiiiiiie over 1 year ago - 1 comment

#146 - About training time cost for pre-train model

Issue - State: open - Opened by CZX-Yui over 1 year ago

#145 - Hello, I used Chinese data for training and found that all the results I got have the first character as "的", such as "的里有一的男士" and "的山小上一的女人和一手子的人". I would like to ask if there is anything else I need to modify besides the data when training Chinese language.

Issue - State: open - Opened by cjt222 almost 2 years ago - 3 comments

#144 - Add 🤗 integration

Pull Request - State: closed - Opened by NielsRogge almost 2 years ago - 1 comment
Labels: cla:signed

#143 - Got 10 same words in respons.

Issue - State: open - Opened by zss977-web almost 2 years ago - 1 comment

#142 - A n error about "The size of tensor a (96) must match the size of tensor b (288) at non-singleton dimension 0"

Issue - State: open - Opened by ZhenyuLiu-SYSU almost 2 years ago - 5 comments

#141 - Visual Question Answering's confidence

Issue - State: open - Opened by kosarkazemi almost 2 years ago - 2 comments

#140 - Why convert images to cpu in BLIP

Issue - State: open - Opened by itsik1 almost 2 years ago

GitHub / salesforce/BLIP issues and pull requests