Issues
- 2
About the lora config alpha and rank
#58 opened by lucasjinreal - 1
No module named dataset
#76 opened by liguopeng0923 - 0
- 1
Eval Results.
#73 opened by liguopeng0923 - 0
- 6
ValueError: The input provided to the model are wrong. The number of image tokens is 0 while the number of image given to the model is 1. This prevents correct indexing and breaks batch generation.
#32 opened by zhangchunjie1999 - 0
Question about the training data
#74 opened by Kendrick-Powehi-Z - 0
Training dataset keys_indexfile
#72 opened by linxid - 1
Training Dataset
#71 opened by linxid - 2
Difference of videochat2 jsons and magic jsons.
#52 opened by OliverHxh - 7
Instruction follow ability is weak
#68 opened by lmx760581375 - 0
High video loss
#70 opened by TonyXuQAQ - 0
[TRAINING DATA]About the number of video_qa in magic_jsons, which is more than the reported number in PLLaVA paper!
#67 opened by YeeShih - 3
How to prepare eval dataset.
#57 opened by linxid - 2
Training loss curve
#66 opened by hukkai - 4
finetuning PLLaVA - IMAGEs for Video Analysis
#28 opened by sureshtmca - 3
- 3
Clarifying Ambiguity in Training Data
#47 opened by patrick-tssn - 0
- 0
What is the prompt used to evaluate models answer on VideoChatGPT generation benchmark?
#64 opened by ADiko1997 - 2
The inference result is all like a mess
#63 opened by lucasjinreal - 4
Inference error
#61 opened by lucasjinreal - 2
How to fintune base on your fintuned result
#56 opened by liuao743 - 0
- 0
Abou the dataset
#59 opened by lucasjinreal - 2
Training time consumption
#51 opened by hmxiong - 8
eval on fintuning model
#50 opened by liuao743 - 0
Multi-GPU Inference for 34B Model
#48 opened by serwansj - 2
How to train only the projector?
#54 opened by gaowei724 - 1
- 0
What's the 7B model score on mvbench?
#53 opened by MonolithFoundation - 3
finetune problem
#43 opened by AshOneN - 5
- 1
Checkpoints for trained models?
#49 opened by serwansj - 0
It seems that the most effective point of increase is to increase the size of llm
#44 opened by xmy0916 - 0
Possible error in doctring for "pixel_values"
#42 opened by tomyoung903 - 1
- 1
Finetune my own's tasks based on pllava-7b, but got "image_to_overwrite.sum() != image_features.shape[:-1].numel()" assertion failure.
#39 opened by gaowei724 - 1
Question about fine-tuning
#23 opened by l-salewski - 1
Bug in tasks/eval/eval_utils.py
#40 opened by patrick-tssn - 1
demo.sh issues in 34b model
#34 opened by Slezge - 2
Dataset Resquest
#36 opened by mingzeG - 0
CLI inference
#37 opened by YepJin - 0
About vram and int-4 version
#35 opened by joseph16388 - 5
- 0
34b can not use gpu
#33 opened by Lvqin001 - 2
Encountered an error while running the demo: the data type (dtype) of the llm does not match the data type of the vision model.
#29 opened by gaowei724 - 0
Missing LICENSE file
#30 opened by bhokaal2k - 3
train_pllava_13b.sh运行时会报错
#26 opened by emmating12 - 1
Cannot run demo.sh with pllava-34b
#25 opened by AmitRozner