Comments (8)
@tuanhui-li Would you mind to indicate which paper? This repo contains multiple searching algorithms.
from autodl-projects.
Oh, I mean the paper of 'Network Pruning via Transformable Architecture Search', in configs/NeurIPS-2019
, where only contains the model config of CIFAR-X. Can you share the model config of ImageNet and the final checkpoints reported on Table 3 of the original paper ? Thanks .
from autodl-projects.
@tuanhui-li Thanks for your interests. I will upload the configs in 24 hours. Currently, I'm largely re-organizing the codes and can not push those configs right now, sorry for the inconvenience.
Regarding the final checkpoints, I find some of them in my server and will upload them soon, but some checkpoints are lost. Thanks for your patience.
from autodl-projects.
@D-X-Y Maybe you can also provide the shell scripts for ImageNet model searching and inference, which will be much helpful. I think these hyper-parameters for ImageNet are different from the CIFARX, and the results on ImageNet are much more important.
from autodl-projects.
@tuanhui-li Sure, please see the following updates:
- ImageNet configs at https://github.com/D-X-Y/AutoDL-Projects/tree/master/configs/NeurIPS-2019
- I double-checked the log and script, I fixed the config-path for ImageNet (details at https://github.com/D-X-Y/AutoDL-Projects/blob/master/scripts/tas-infer-train.sh#L32)
- Logs for searching and training(you-called-inference) on CIFAR: https://drive.google.com/open?id=1h0RPrbXL-79U-CBos1wOst60oNxOkrWS
- Logs for training the searched shape of ResNet-18 on ImageNet: https://drive.google.com/open?id=1FTXnbKnOcwsKVCQvNUgU6RVqYtRiiHMw
- Logs for training the searched shape of ResNet-50 on ImageNet: https://drive.google.com/open?id=1TwCBgbFzGGMbdglLWJQPyMZgIm9acRaz
- I did not find the searching log of ImageNet on my current server. I will try to look at other devices.
Hope my codes and checkpoints are useful to you :)
from autodl-projects.
@D-X-Y I am sorry to bother you again ! Can you release the script for searching imagenet model ?
from autodl-projects.
Hi, I am also looking forward to search scripts of ResNet 18/50 on Imagenet dataset. Thank you very much!
from autodl-projects.
Sorry for the late reply. I will try to find
from autodl-projects.
Related Issues (20)
- Fail to generate the onnx file HOT 2
- ENAS in NATS-Bench is better than that in NAS-Bench-201
- The link of NAS-BENCH-201-4-v1.0-archive.tar failed HOT 2
- How to quickly retrain network architecture from scratch after getting the best topology HOT 2
- Where can I get the code for regularized_ea train and evaluation HOT 2
- Difference between GDAS variants HOT 3
- failed to download NAS-Bench-201-v1_1-096897.pth HOT 3
- https://github.com/D-X-Y/AutoDL-Projects/blob/main/docs/NATS-Bench.md HOT 1
- Loading weights for NAS-Bench 201 HOT 5
- Usage of algorithm
- Where is the documentation
- Query related to reduce_concat and normal_concat in Genotype
- training time on NATS-Benchmark HOT 1
- NASBench201 dataset difficult to download HOT 1
- how to decompress .bpzip file HOT 1
- Question regarding NAS-Bench-201
- AutoDL algorithm HOT 3
- Questions about GDAS sampling process
- Regarding Flop and Parameter count in NASBench201 cifar10 dataset
- Question for create the network from api in nas_bench_201 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 autodl-projects.