issues Search Results · repo:facebookresearch/perception_models language:"Jupyter Notebook"
Filter by
67 results
(64 ms)67 results
infacebookresearch/perception_models (press backspace or delete to remove)Hi!
Thank you for the amazing work.
I tried to reproduce the results of PLM-3B on RefCOCO. After downloading the checkpoints from HF, I use the prompt as in
https://github.com/facebookresearch/perception_models/blob/main/apps/plm/notebook_demos/image_grounding.ipynb ...
zgq1879
- Opened 3 days ago
- #98
Hi, sorry if this is a dumb question, but is there a way to use this repo with python version =3.10? Or is there some
crucial dependency on 3.12? The environment I need to work on does not support anything ...
ssamt
- Opened 4 days ago
- #97
Hello, first of all, thank you for your work and making it open source. Looking at the linear_probe.py file, I see that
you have implemented an hyper parameters sweep approach to find the best weight-decay ...
Antuke
- Opened 25 days ago
- #96
Hello, I have checked Appendix C.3 for the evaluation protocol on PLM-VideoBench. The prompt you use to Llama to
evaluate the region caption is written as the following: Your task is to compare a given ...
MSiam
- Opened on Sep 22
- #95
[2025-09-08T17:32:09+08:00] videoac-plm-7qzvs File /eval/videoac/lmms-eval/lmms_eval/models/simple/plm.py , line 68, in
init [2025-09-08T17:32:09+08:00] videoac-plm-7qzvs model, tokenizer, config =
load_consolidated_model_and_tokenizer(cfg.pretrained) ...
zhangyx0417
- Opened on Sep 8
- #93
yuchenlichuck
- Opened on Aug 23
- #88
Hello,
The Perception Encoder checkpoints look excellent—thank you for releasing them! However, I noticed that only the ViT-L/G
variants are available, and there are no H or g variants. This leaves a ...
JihwanEom
- Opened on Aug 21
- #87
Summary
A boolean mask derived from text positions is incorrectly applied to the flattened image features during stitching,
causing a shape mismatch and a runtime crash during generation.
Impact: blocks ...
bug
PLM
EladSharony
- 1
- Opened on Aug 14
- #86
Hi, so I had some questions after reading the paper.
Unlike the video dataset, I could find very little information about the image-text dataset used in pretraining ( 2.3B
image-text dataset curated using ...
PE
- 4
- Opened on Aug 12
- #85
When I installed this repo by pip install git+https://github.com/facebookresearch/perception_models.git, the file
core/vision_encoder/bpe_simple_vocab_16e6.txt.gz was not downloaded. Manually downloading ...
- 4
- Opened on Aug 10
- #84
Learn how you can use GitHub Issues to plan and track your work.
Save views for sprints, backlogs, teams, or releases. Rank, sort, and filter issues to suit the occasion. The possibilities are endless.Learn more about GitHub IssuesProTip!
Press the /
key to activate the search input again and adjust your query.Learn how you can use GitHub Issues to plan and track your work.
Save views for sprints, backlogs, teams, or releases. Rank, sort, and filter issues to suit the occasion. The possibilities are endless.Learn more about GitHub IssuesProTip!
Press the /
key to activate the search input again and adjust your query.