Comments (3)
Hi, the first language model that we used to build MotionGPTs is LLaMA-13B. However, it shows insufficient performance and low training efficiency. We assume the reason is the limited dataset size compared to the large parameters and language data of LLaMA.
Then, we thus choose T5-770M, a small but common language model, as our final backbone, because many previous vision-language multimodal works, like Unified-IO and BLIP, have chosen T5, this encoder-decoder architecture. It shows a strong power to address multi-modal tasks. In addition, the decoder-only model has the advantage for self-supervised without pair data while we have paired data which this advance is greatly weakened. We are still working on collecting a large motion dataset for larger motion-language models.
We have evaluated MotionGPT on GPT-2 and are working on LLaMA-2+LORA. Please refer to the below.
from motiongpt.
Did you only do fine-tuning, or did you also perform pre-training?
from motiongpt.
Did you only do fine-tuning, or did you also perform pre-training?
Hello @ChangeNext
We employ both pre-training and fine-tuning processes for the T5 and GPT-2 models to ensure they are optimally adapted for our specific tasks.
from motiongpt.
Related Issues (20)
- How to reproduce the motion complete result in the teaser video?
- If I only want to evaluate this motionGPT, do I need to build Humanml3D as well HOT 1
- Motion2Text generation always gives something irrelevant
- Question on m2t-motionGPT evaluation metrics HOT 2
- RuntimeError: The expanded size of the tensor (3) must match the existing size (263) at non-singleton dimension 1. Target sizes: [22, 3]. Tensor sizes: [263] HOT 2
- the demo page gives a runtime error HOT 2
- question about the task
- How can I get the 2d joints from the SMPL mesh nptyfile HOT 1
- How to save the VQVAE's weight separately from the whole model? HOT 1
- Training Kit-ML encountered dimension mismatch problem HOT 1
- size mismatch for main.0.weight HOT 1
- Camera location includes nan values. HOT 2
- safetensors_rust.SafetensorError: Error while deserializing header: HeaderTooLarge HOT 4
- Should the END-EPOCH parameter be set to 999999? HOT 3
- Issue with Training MotionGPT on Multiple Devices
- Issue with Increasing batchSize in Visualization Part2 Script: Create SMPL meshes with mulit batch
- Training on custom Dataset HOT 3
- Link for SMPL models not working HOT 1
- Hugging face demo not working required for comparison on Motion2Text
- Gradio: No supported video format or MIME type found 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 motiongpt.