Ecosyste.ms: Issues

An open API service for providing issue and pull request metadata for open source projects.

GitHub / tri-ml/prismatic-vlms issues and pull requests

#49 - generate.py not working

Issue - State: open - Opened by ashmalvayani 3 months ago

#49 - generate.py not working

Issue - State: open - Opened by ashmalvayani 3 months ago

#47 - AttributeError: module 'torch.distributed' has no attribute 'Logger'

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

#45 - Save normalizers

Pull Request - State: closed - Opened by siddk 6 months ago

#45 - Save normalizers

Pull Request - State: closed - Opened by siddk 6 months ago

#44 - Custom Vision Encoder and LLM backbone

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

#44 - Custom Vision Encoder and LLM backbone

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

#43 - Using multiple GPUs for inference

Issue - State: open - Opened by yunbinmo 8 months ago - 1 comment

#43 - Using multiple GPUs for inference

Issue - State: open - Opened by yunbinmo 8 months ago - 1 comment

#41 - Code for Creating VLMs using New Language Backbones

Pull Request - State: open - Opened by RylanSchaeffer 8 months ago - 2 comments

#41 - Code for Creating VLMs using New Language Backbones

Pull Request - State: open - Opened by RylanSchaeffer 8 months ago - 2 comments

#40 - One question about running inference

Issue - State: open - Opened by Li-private 8 months ago - 4 comments

#40 - One question about running inference

Issue - State: open - Opened by Li-private 8 months ago - 4 comments

#39 - Multinode training

Issue - State: open - Opened by budzianowski 8 months ago - 3 comments

#39 - Multinode training

Issue - State: open - Opened by budzianowski 8 months ago

#38 - move barrier to before saving checkpoint to reduce timeouts when saving

Pull Request - State: open - Opened by jensen-gao 8 months ago - 1 comment

#37 - Revert "OpenVLA Release"

Pull Request - State: closed - Opened by siddk 8 months ago

#37 - Revert "OpenVLA Release"

Pull Request - State: closed - Opened by siddk 8 months ago

#36 - OpenVLA Release

Pull Request - State: closed - Opened by siddk 8 months ago

#36 - OpenVLA Release

Pull Request - State: closed - Opened by siddk 8 months ago

#35 - `clip_grad_norm_` in fsdp

Issue - State: closed - Opened by zeyuanyin 9 months ago - 1 comment

#35 - `clip_grad_norm_` in fsdp

Issue - State: closed - Opened by zeyuanyin 9 months ago - 1 comment

#34 - `unpack_tuple()` is no longer correct with timm v1.0.3

Issue - State: open - Opened by yukw777 9 months ago - 2 comments

#34 - `unpack_tuple()` is no longer correct with timm v1.0.3

Issue - State: open - Opened by yukw777 9 months ago - 2 comments

#33 - Questions about mixed-precision training for visual backbones

Issue - State: closed - Opened by tayton42 9 months ago - 3 comments

#33 - Questions about mixed-precision training for visual backbones

Issue - State: closed - Opened by tayton42 9 months ago - 2 comments

#32 - The reproduction issue of DINOv2 + SigLIP 384px (Naive Resize)

Issue - State: closed - Opened by tayton42 9 months ago - 12 comments

#32 - The reproduction issue of DINOv2 + SigLIP 384px (Naive Resize)

Issue - State: closed - Opened by tayton42 9 months ago - 12 comments

#30 - question about training data

Issue - State: closed - Opened by fyting 10 months ago - 1 comment

#30 - question about training data

Issue - State: closed - Opened by fyting 10 months ago - 1 comment

#29 - question about vit's fsdp wrapping policy

Issue - State: closed - Opened by lukaemon 10 months ago - 2 comments

#29 - question about vit's fsdp wrapping policy

Issue - State: closed - Opened by lukaemon 10 months ago - 2 comments

#27 - Add New LLM Backbones

Pull Request - State: closed - Opened by siddk 10 months ago

#27 - Add New LLM Backbones

Pull Request - State: closed - Opened by siddk 10 months ago

#26 - Cleanup and Lint, add 224px Prism Models

Pull Request - State: closed - Opened by siddk 10 months ago

#26 - Cleanup and Lint, add 224px Prism Models

Pull Request - State: closed - Opened by siddk 10 months ago

#25 - Need help when incorporating LLama3

Issue - State: closed - Opened by Hannibal046 10 months ago - 1 comment

#25 - Need help when incorporating LLama3

Issue - State: closed - Opened by Hannibal046 10 months ago - 1 comment

#24 - Update README.md

Pull Request - State: closed - Opened by SamuelSchmidgall 10 months ago - 1 comment

#24 - Update README.md

Pull Request - State: closed - Opened by SamuelSchmidgall 10 months ago - 1 comment

#22 - Why second to last layer Vision Transformer features?

Issue - State: closed - Opened by gorjanradevski 10 months ago - 1 comment

#22 - Why second to last layer Vision Transformer features?

Issue - State: closed - Opened by gorjanradevski 10 months ago - 1 comment

#21 - Can I training your code with V100 GPU

Issue - State: closed - Opened by swordlidev 10 months ago - 1 comment

#21 - Can I training your code with V100 GPU

Issue - State: closed - Opened by swordlidev 10 months ago - 1 comment

#20 - How to finetune starting from a Prismatic VLM checkpoint

Issue - State: closed - Opened by djghosh13 10 months ago - 1 comment

#20 - How to finetune starting from a Prismatic VLM checkpoint

Issue - State: closed - Opened by djghosh13 10 months ago - 1 comment

#19 - Feature request

Issue - State: closed - Opened by lucasjinreal 10 months ago - 1 comment

#19 - Feature request

Issue - State: closed - Opened by lucasjinreal 10 months ago - 1 comment

#18 - Love your code!

Issue - State: closed - Opened by Hannibal046 10 months ago - 1 comment

#18 - Love your code!

Issue - State: closed - Opened by Hannibal046 10 months ago - 1 comment

#17 - Multi-dataset support issue.

Issue - State: closed - Opened by tayton42 10 months ago - 1 comment

#17 - Multi-dataset support issue.

Issue - State: closed - Opened by tayton42 10 months ago - 1 comment

#16 - Llava / Prismatic with LoRA

Issue - State: open - Opened by gorjanradevski 10 months ago - 6 comments

#16 - Llava / Prismatic with LoRA

Issue - State: open - Opened by gorjanradevski 10 months ago - 6 comments

#15 - training error

Issue - State: closed - Opened by tayton42 10 months ago - 2 comments

#15 - training error

Issue - State: closed - Opened by tayton42 10 months ago - 2 comments

#14 - Inconsistent API for Vision Backbones?

Issue - State: closed - Opened by RylanSchaeffer 10 months ago - 13 comments

#14 - Inconsistent API for Vision Backbones?

Issue - State: closed - Opened by RylanSchaeffer 10 months ago - 13 comments

#13 - Does this software support CogVLM?

Issue - State: closed - Opened by PhilipAmadasun 11 months ago - 1 comment

#13 - Does this software support CogVLM?

Issue - State: closed - Opened by PhilipAmadasun 11 months ago - 1 comment

#12 - Quantization support

Issue - State: open - Opened by show981111 11 months ago - 2 comments

#12 - Quantization support

Issue - State: open - Opened by show981111 11 months ago - 2 comments

#11 - Any plan to support "Dynamic High Resolution" proposed in LLaVA v.16?

Issue - State: closed - Opened by yushuinanrong 11 months ago - 1 comment

#10 - Do you have plan to add SAM as a visual encoder?

Issue - State: open - Opened by StarCycle 11 months ago - 5 comments

#8 - Training a one-stage model

Issue - State: closed - Opened by shikhar-srivastava 11 months ago - 3 comments

#6 - Adding support for other base LLMs

Issue - State: closed - Opened by shikhar-srivastava 12 months ago - 7 comments

#5 - evaluatio scripts

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

#4 - Have you tried only dinov2 as vision encoder?

Issue - State: closed - Opened by LinB203 12 months ago - 1 comment

#3 - Does it support input format of multiple images + text?

Issue - State: closed - Opened by swj0419 about 1 year ago - 2 comments

#2 - pretrain.py won't run with arguments

Issue - State: closed - Opened by rpgrainger-ai about 1 year ago - 2 comments

#1 - How to extract the last hidden outputs from the models?

Issue - State: closed - Opened by swj0419 about 1 year ago - 3 comments