Comments (3)
The data are introduced in Appendix A. I will introduce more details about this part in a revised version (maybe updated in two days).
from cogview.
Can you explain the training resource? It seems omitted in the paper.
We collected about 30 million text-image pairs from multiple channels, and built a 2.5TB new
dataset (after the data was processed into tokens, the size becomes about 250GB). The data are an
extension of project WudaoCorpora
About 50% of the text is in English, including Conceptual
Captions [42]. They are translated into Chinese by machine translation. In addition, we did not
remove the watermarks and white edges in the dataset even though they affect the quality of generated
images, because we think it will not influence the conclusions of our paper from the perspective of
research.
from cogview.
Hi @gaopengcuhk. We have updated the introduction of data in the Appendix A of the paper, which is also quoted by @johnpaulbin .
from cogview.
Related Issues (20)
- fused_layer_norm_cuda HOT 1
- Can not download Wudao-Wenhui
- Why BAAI pretrained models can't be downloaded directly?
- How to set style in the code? HOT 1
- pretrain and finetune loss and lr
- ImportError: cannot import name 'UnencryptedCookieSessionFactoryConfig' from 'pyramid.session' (unknown location) on google colab HOT 1
- RuntimeError: CUDA error: CUBLAS_STATUS_EXECUTION_FAILED when calling `cublasGemmEx( handle, opa, opb, m, n, k, &falpha, a, CUDA_R_16F, lda, b, CUDA_R_16F, ldb, &fbeta, c, CUDA_R_16F, ldc, CUDA_R_32F, CUBLAS_GEMM_DFALT_TENSOR_OP)
- Wudao down? Or the host was changed? HOT 3
- Is the CogView2 demo avaliable here the same as the CogView2 paper released? HOT 1
- Do you have a model checkpoint smaller than 7GB?
- model parallel train
- Bug:The following error is reported after executing the command “./scripts/pretrain_single_node.sh --debug”.How can I solve this problem? HOT 1
- After training, there are only 2G Pt files left. What's wrong with missing files? HOT 1
- Max Text Length HOT 1
- Copyright problems HOT 1
- How to use the pytorch_model.bin file downloaded from WuDao?
- Wudao-Wenhui(悟道-文汇)302 HOT 1
- How to resume? HOT 2
- CogView-fashion
- Question for bird-and-animal training
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 cogview.