Comments (7)
You can try this: https://github.com/PaddlePaddle/PASSL/blob/main/docs/Train_CLIP_model.md
The link includes a demo which using texts to retrieve images.
from passl.
You can try this: https://github.com/PaddlePaddle/PASSL/blob/main/docs/Train_CLIP_model.md
The link includes a demo which using texts to retrieve images.
thanks, and i want to use paddlepaddle/moco v2 (https://github.com/PaddlePaddle/PASSL/blob/main/configs/moco/moco_v2_r50.yaml)to do image similarity matching experiments locally(linux or mac), can you give me some suggestions
from passl.
First of all, you can download paddle moco v2 model from this link .
Second, you should extract the moco v2 backbone weights following the step 2 in our MoCo doc .
Last but not least, you can run your image similarity matching experiments following our clip demo.
from passl.
AI Studio Project: https://aistudio.baidu.com/aistudio/projectdetail/2332016?_=1630569049096
from passl.
AI Studio Project: https://aistudio.baidu.com/aistudio/projectdetail/2332016?_=1630569049096
buddy, your update is very timely. I have been using the clip demo just now to encounter'' image_input.npy ''(github), which solved this, but it also prompted ValueError: paddle.load
can not parse the file:ViT-B-32.pdparams .I am running on a mac.I just want to experiment with moco, it's too torturous
from passl.
AI Studio Project: https://aistudio.baidu.com/aistudio/projectdetail/2332016?_=1630569049096
buddy, your update is very timely. I have been using the clip demo just now to encounter'' image_input.npy ''(github), which solved this, but it also prompted ValueError:
paddle.load
can not parse the file:ViT-B-32.pdparams .I am running on a mac.I just want to experiment with moco, it's too torturous
paddle version:2.1.2
from passl.
Do you install passl correctively? or download vit-b-32 model?
Make sure the code bellow is running fluently:
pip install passl==0.0.4 -U -i https://pypi.tuna.tsinghua.edu.cn/simple
pip install scikit-image -U -i https://pypi.tuna.tsinghua.edu.cn/simple
wget https://passl.bj.bcebos.com/models/ViT-B-32.pdparams
from passl.
Related Issues (20)
- [FEATURE] Plan for DINO HOT 1
- trainer没有保存权重的设置吗 HOT 1
- 后续会有更多预训练模型权重吗? HOT 1
- BEiT配置yaml配置文件有问题 HOT 1
- Unintended behavior in clip_logit_scale HOT 1
- 求MAE的预训练参数 HOT 2
- ConvNext的实现为什么要自定义LayerNorm? HOT 1
- 训练完没有权重
- Does the passl model support c++ deployment? HOT 1
- 建议在repo的简介中加入英文版本 HOT 1
- BYOL的预训练中好像使用了gt_label? HOT 1
- 只有正样本,没有负样本,这个框架能用在这样的场景吗 HOT 2
- KeyError: "No object named 'MAE_FINETUNE' found in 'MODEL' registry!" HOT 1
- bug HOT 2
- dinov2 pretrain HOT 1
- 关于DINO/DINOV2的支持 HOT 1
- An error when I run SWAV: HOT 2
- PASSL训练大模型 HOT 1
- 如何训练自有数据集? HOT 1
- ModuleNotFoundError: No module named 'passl.engine.engine' HOT 1
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from passl.