Ecosyste.ms: Issues
An open API service for providing issue and pull request metadata for open source projects.
GitHub / cambrian-mllm/cambrian issues and pull requests
#79 - How to plot images?(cambrian7m.png)
Issue -
State: open - Opened by banduoba about 1 month ago
#78 - hotfix: rm comments
Pull Request -
State: closed - Opened by ellisbrown about 2 months ago
#77 - MLLM Eval
Pull Request -
State: closed - Opened by ellisbrown 2 months ago
#76 - RuntimeError: Cannot set version_counter for inference tensor
Issue -
State: open - Opened by anguoyuan 2 months ago
#75 - Fail to Reproduce the Performance on MMVP.
Issue -
State: closed - Opened by Haochen-Wang409 2 months ago
- 2 comments
#74 - AttributeError: module 'torch.nn.functional' has no attribute 'scaled_dot_product_attention'. Did you mean: '_scaled_dot_product_attention'?
Issue -
State: open - Opened by yfpeng1234 2 months ago
#73 - Cambrian Data: Sources unclear in the JSONs
Issue -
State: open - Opened by kushalkafle 3 months ago
#72 - Questions about cross attention during performing SVA.
Issue -
State: closed - Opened by liujpg 3 months ago
- 4 comments
#71 - Alignment data seems broken
Issue -
State: open - Opened by je1lee 3 months ago
- 2 comments
#70 - GPU memory usage increases at each inference
Issue -
State: open - Opened by FSet89 3 months ago
#69 - What does the `source` mean in the instruction tuning jsonl data?
Issue -
State: closed - Opened by daa233 3 months ago
- 1 comment
#68 - Generation after wrapping with XLA FSDP using TPU
Issue -
State: open - Opened by pUmpKin-Co 3 months ago
#67 - Support for multi-gpu Pre-training Code
Issue -
State: open - Opened by Nipun0309 4 months ago
- 2 comments
#66 - Evaluation with datasets that contains multiple images
Issue -
State: closed - Opened by Nyandwi 4 months ago
- 2 comments
#65 - RADIOv2.5 as vision encoder
Issue -
State: open - Opened by gheinrich 4 months ago
#64 - Support GPU Training
Pull Request -
State: open - Opened by TideDra 4 months ago
- 2 comments
#63 - 拉了一个多模态大模型技术交流群,大家可以加入进来进行技术交流
Issue -
State: closed - Opened by feihuamantian 4 months ago
#62 - Cannot run offline only with model file
Issue -
State: open - Opened by Rising0321 4 months ago
- 1 comment
#61 - For experiments in section 3.3, did you use the cambrian model architecture or llava?
Issue -
State: closed - Opened by xrc10 4 months ago
- 2 comments
#60 - Model instruction following ability is very weak
Issue -
State: open - Opened by linwenzhao1 4 months ago
#59 - [BUG] Unable to download model weights from Huggingface
Issue -
State: closed - Opened by peiqi-liu 4 months ago
- 2 comments
#58 - Cannot run inference with Cambrian-1-34B on multi-GPU
Issue -
State: open - Opened by Lopa07 4 months ago
- 3 comments
#57 - More details about cambrain-1
Issue -
State: closed - Opened by LoverLost 4 months ago
- 1 comment
#56 - There is an error in Table 3, Vision Backbone in lines 3 and 6 are both "SigLIP+DINOv2+ConvNext"
Issue -
State: closed - Opened by dingangui 4 months ago
- 1 comment
#55 - Validation Data in Cambrian Dataset
Issue -
State: open - Opened by SaraGhazanfari 4 months ago
#54 - Hugging Face Demo
Issue -
State: open - Opened by andupotorac 4 months ago
#53 - where is the sva projector?
Issue -
State: closed - Opened by ZhuGeKongKong 4 months ago
- 1 comment
#52 - Similarities between Cambrian-1 and MouSi
Issue -
State: open - Opened by guitaowufeng 4 months ago
#51 - Experiment on Cambiran7m with system prompt and Minigemini data, former got even worse result
Issue -
State: open - Opened by MonolithFoundation 4 months ago
- 2 comments
#50 - An error when using Cambrian7M with system prompts
Issue -
State: open - Opened by LiWentomng 4 months ago
#49 - Using Cambrian 7M as instruction data got extremly bad result
Issue -
State: open - Opened by lucasjinreal 4 months ago
- 5 comments
#48 - different cuda version
Issue -
State: open - Opened by sgmzhou4 4 months ago
#47 - Cambrian-phi3-3b
Issue -
State: closed - Opened by BrenchCC 4 months ago
- 1 comment
#46 - [Bug] Concerns about data quality: Very serious inconsistencies between graphics and text.
Issue -
State: open - Opened by vikyzeng 4 months ago
- 2 comments
#45 - Inference demo for pretrained checkpoints
Issue -
State: open - Opened by xunannancy 4 months ago
#43 - ConnectionResetError: [Errno 104] Connection reset by peer
Issue -
State: open - Opened by Alexbalala 4 months ago
#42 - UnicodeDecodeError: 'gbk' codec can't decode byte 0x92 in position 102: illegal multibyte sequence
Issue -
State: open - Opened by prettyfaceYAO 4 months ago
#41 - When using inference.py I meet this package problem.Could this be related to changing the torch version to 2.0.1 for CUDA 11.7?
Issue -
State: closed - Opened by Chengnotwang 4 months ago
- 5 comments
#40 - [BUG] I think there are some data's image are not correctly set.
Issue -
State: open - Opened by Luodian 4 months ago
#39 - Update README.md
Pull Request -
State: closed - Opened by tsb0601 4 months ago
#38 - Question about data_engine part dataset
Issue -
State: closed - Opened by vikyzeng 4 months ago
- 1 comment
#37 - Question about Cambrian-Alignment dataset
Issue -
State: open - Opened by HYLcool 4 months ago
- 5 comments
#36 - data_engine file cannot be decompressed
Issue -
State: closed - Opened by LiWentomng 4 months ago
- 1 comment
#35 - MultiKVCrossAttention and VisionCrossAttentionLayer forwards do not match
Issue -
State: closed - Opened by wamiq-reyaz 4 months ago
- 1 comment
#34 - Multi image inference
Issue -
State: open - Opened by Beakboomboom 4 months ago
- 2 comments
#33 - sync data engine with main
Pull Request -
State: closed - Opened by ellisbrown 4 months ago
#32 - NETWORK ERROR DUE TO HIGH TRAFFIC. PLEASE REGENERATE OR REFRESH THIS PAGE.
Issue -
State: open - Opened by Iven2132 4 months ago
- 2 comments
#31 - nyu-visionx/cambrian-phi3-3b benchmarks
Issue -
State: closed - Opened by danieltudosiu 4 months ago
- 3 comments
#30 - Finetune with GPU
Issue -
State: open - Opened by haluk-pointr 4 months ago
#29 - Modify config.json or code for downloaded vision encoder model?
Issue -
State: open - Opened by NiYueLiuFeng 4 months ago
#28 - add CV-Bench announcement
Pull Request -
State: closed - Opened by ellisbrown 4 months ago
#27 - Running the inference.py program reported an error
Issue -
State: closed - Opened by hyyuan123 4 months ago
- 1 comment
#26 - High resolution images
Issue -
State: closed - Opened by Shade5 4 months ago
- 1 comment
#25 - Error in inference.py when multiple GPUs are available. [BUG]
Issue -
State: open - Opened by ZeenSong 4 months ago
- 4 comments
Labels: bug
#24 - chore: update train_fsdp.py
Pull Request -
State: closed - Opened by eltociear 4 months ago
- 1 comment
#23 - Question About Table 4
Issue -
State: closed - Opened by homiec 4 months ago
- 2 comments
Labels: question
#22 - fix bug in query initialization
Pull Request -
State: closed - Opened by penghao-wu 4 months ago
#21 - Minimum vram to safely run?
Issue -
State: open - Opened by KSimulation 4 months ago
- 1 comment
#20 - 【bug】do_sample= False still get random output
Issue -
State: open - Opened by CSEEduanyu 4 months ago
- 11 comments
#19 - why cam34b weight sum 130G?
Issue -
State: closed - Opened by CSEEduanyu 4 months ago
- 4 comments
Labels: question
#18 - Dataengine
Pull Request -
State: closed - Opened by Charithaakula 4 months ago
#17 - HF transformers support
Issue -
State: open - Opened by Iven2132 4 months ago
- 4 comments
#16 - Where is the Cambrian 7M?
Issue -
State: closed - Opened by lucasjinreal 4 months ago
#14 - Can you release the 161K science-related data separately?
Issue -
State: closed - Opened by Weiyun1025 5 months ago
- 2 comments
Labels: enhancement
#13 - Adding Data Engine
Pull Request -
State: closed - Opened by Charithaakula 5 months ago
#12 - 【bug】can not load cambrian-34b
Issue -
State: open - Opened by CSEEduanyu 5 months ago
- 17 comments
Labels: bug
#11 - Question about Figure 9
Issue -
State: closed - Opened by digbangbang 5 months ago
- 2 comments
Labels: question
#10 - Question about figure 7
Issue -
State: closed - Opened by digbangbang 5 months ago
- 1 comment
#9 - Can it accept a video as input?
Issue -
State: closed - Opened by LinaZhangCoding 5 months ago
- 1 comment
#8 - [Question about paper] The number of visual tokens
Issue -
State: closed - Opened by terry-for-github 5 months ago
- 1 comment
#7 - Configurations of different size of cambrian models
Issue -
State: closed - Opened by junming-yang 5 months ago
- 2 comments
#6 - support Chinese?
Issue -
State: closed - Opened by lorashen 5 months ago
- 1 comment
#5 - there is an error with 'llava_llama'
Issue -
State: closed - Opened by zhangwei730 5 months ago
- 2 comments
#4 - Support Of Multiple Images Infering
Issue -
State: closed - Opened by caichuang0415 5 months ago
- 1 comment
#3 - Internet Data Engine publish
Issue -
State: closed - Opened by HyunHo99 5 months ago
- 2 comments
Labels: enhancement
#2 - Dataset License
Issue -
State: closed - Opened by JetGang 5 months ago
- 2 comments
#1 - Gradio demo working
Pull Request -
State: closed - Opened by ellisbrown 5 months ago