Comments (2)
feihugis
commented on May 19, 2024
Have you guys looked at fairseq eval_lm?
It's not generative, but I was wondering whether any of these tricks would work.
Thanks in advance!
@sshleifer I have run a quick test using the below command with/without FastSeq, but did not see big speed difference. I think this is as expected after the initial investigation. The reasons are that eval_lm did not have the duplication issues in the cached key and value as the forward process just runs once; the no n-gram block operation has not been used; and the asynchronized pipeline with parallel IO is not involved.
python eval_lm.py adaptive_lm_wiki103.v2/data-bin/ --path adaptive_lm_wiki103.v2/model.pt --sample-break-mode complete --max-tokens 3072 --context-window 2560 --softmax-batch 1024
from fastseq.
feihugis
commented on May 19, 2024
Close this issue. Please feel free to reopen it if there are more questions.
from fastseq.
Related Issues (20)
- Does fastseq support cpu HOT 3
- NMT models speedup abnormally related to batch size HOT 2
- Can the fastseq install on windows? HOT 3
- Where to read EL-Attention source code for huggingface-transformers HOT 4
- In which file to read the source code implementation of El-Attention for self-attention HOT 1
- EL-attention GPT-2 HOT 2
- Running error with PyTorch 1.12.1 HOT 2
- fairseq/transformers unit test modify local environment HOT 3
- Transformers unit tests failure
- ACTION REQUIRED: Microsoft needs this private repository to complete compliance info
- RuntimeError: CUDA error: no kernel image is available for execution on the device HOT 7
- ModuleNotFoundError: No module named 'fastseq.models' HOT 2
- Does it support Tensorflow 2? HOT 3
- Any end-end inference example with Google Colab & HuggingFace HOT 2
- Does it support model seq2seq with encoder, and decoder base on lstm, bi-lstm? HOT 2
- Support for ONNX models & INT8 quantization HOT 2
- Support for HF's transformers 3.1+ HOT 6
- Support for Hugging Face's PEGASUS Model HOT 6
- Support for current fairseq 0.10.2 HOT 13
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 fastseq.