Issues
- 0
How to get the fine-tuned gpt2?
#87 opened by FelixFeiyu - 1
Need help with installation
#82 opened by sashanksaya - 0
The loss curve during clipcap training
#85 opened by gWeiXP - 1
Prefix Interpretability
#84 opened by muhammad-azmain-mahtab - 1
About inference
#83 opened by szxmachinelearning - 4
- 1
- 1
clip_prefix_captioning_inference
#80 opened by Itay89 - 7
- 1
Question About the ClipCocoDataset in train.py
#34 opened by ycchanau - 0
- 4
have you tried different CLIP models?
#53 opened by dhansmair - 1
- 0
clipcap checkpoints file
#77 opened by SZW-zjbd - 3
Error in Load model weights
#76 opened by inakanishi - 3
Not able to train custom data.
#48 opened by sanjaygunda13 - 8
Some questions about fine-tune with custom dataset
#56 opened by eeyrw - 2
Metrics of ClipCap's Original Performance
#65 opened by chmorfop - 1
- 0
- 3
- 0
How to do eval, how to set the prompt
#73 opened by ASENNIU - 6
- 0
Where is the file 'model_wieghts.pt' exists?
#72 opened by Etty-Cohen - 0
data json
#71 opened by rongtongxueya - 2
Evaluation script for COCO
#39 opened by Eladhi - 3
use different encoder
#61 opened by thesuncool - 2
- 1
Train costom data
#64 opened by zzk2021 - 2
- 3
I would like to ask how can Clip-Cap generate multiple different sentences for one image?
#51 opened by linhuixiao - 7
inference
#43 opened by Twilighter9527 - 1
Parsing conceptual caption does not function properly as it removes some images and replaces them with zero tensor.
#60 opened by mmderakhshani - 0
use multiple gpus to train
#68 opened by one-molamola - 5
model overfitting issue
#59 opened by alchemz - 12
Evaluate problem
#33 opened by Twilighter9527 - 0
- 0
it would be None
#54 opened by enes3774 - 0
- 0
Low variety of outputs.
#50 opened by hideosnes - 1
Has this paper been accepted yet? thank you
#49 opened by linhuixiao - 1
For the transformer (without fine-tuning GPT-2) we provide COCO pretrained model.
#46 opened by fido20160817 - 1
purpose of "id" element in coco train data?
#44 opened by stevebottos - 1
prefix size for Vit/ResNet
#45 opened by scfrank - 4
FInal token index when training a model
#42 opened by robertodessi - 2
Incorrect use of labels for GPT2?
#41 opened by soumyasanyal - 0
Docker run: {"detail":"Not Found"}
#40 opened by Wheest - 7
Reproducing validation results
#38 opened by JHLee0513 - 1
'CLIP' object has no attribute 'clip_project'
#37 opened by AhmedEwis - 0
labels in MLP+GPT2 tuning?
#35 opened by ScottishFold007