update README.md

configs/nas/mmcls/onceforall/README.md

@@ -0,0 +1,49 @@
+# Once-For-All
+
+> [ONCE-FOR-ALL: TRAIN ONE NETWORK AND SPE- CIALIZE IT FOR EFFICIENT DEPLOYMENT](https://arxiv.org/abs/1908.09791)
+
+<!-- [ALGORITHM] -->
+
+## Abstract
+
+We address the challenging problem of efficient inference across many devices and resource constraints, especially on edge devices. Conventional approaches either manually design or use neural architecture search (NAS) to find a specialized neural network and train it from scratch for each case, which is computationally prohibitive (causing CO2 emission as much as 5 cars’ lifetime Strubell et al. (2019)) thus unscalable. In this work, we propose to train a once-for-all (OFA) network that supports diverse architectural settings by decoupling training and search, to reduce the cost. We can quickly get a specialized sub-network by selecting from the OFA network without additional training. To efficiently train OFA networks, we also propose a novel progressive shrinking algorithm, a generalized pruning method that reduces the model size across many more dimensions than pruning (depth, width, kernel size, and resolution). It can obtain a surprisingly large number of sub- networks (> 1019) that can fit different hardware platforms and latency constraints while maintaining the same level of accuracy as training independently. On diverse edge devices, OFA consistently outperforms state-of-the-art (SOTA) NAS methods (up to 4.0% ImageNet top1 accuracy improvement over MobileNetV3, or same accuracy but 1.5× faster than MobileNetV3, 2.6× faster than EfficientNet w.r.t measured latency) while reducing many orders of magnitude GPU hours and CO2 emission. In particular, OFA achieves a new SOTA 80.0% ImageNet top-1 accuracy under the mobile setting (\<600M MACs). OFA is the winning solution for the 3rd Low Power Computer Vision Challenge (LPCVC), DSP classification track and the 4th LPCVC, both classification track and detection track.
+
+## Introduction
+
+We product inference models which are published by official Once-For-All repo and converted by MMRazor.
+
+### Subnet test on ImageNet
+
+```bash
+sh tools/slurm_test.sh $PARTITION $JOB_NAME \
+  configs/nas/mmcls/onceforall/ofa_mobilenet_subnet_8xb256_in1k.py \
+  $STEP2_CKPT --work-dir $WORK_DIR --eval accuracy
+```
+
+## Results and models
+
+| Dataset  |       Supernet       |                                  Subnet                                  | Params(M) | Flops(G) | Top-1 |                           Config                            |                                                                                        Download                                                                                        |         Remarks         |
+| :------: | :------------------: | :----------------------------------------------------------------------: | :-------: | :------: | :---: | :---------------------------------------------------------: | :------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------: | :---------------------: |
+| ImageNet | AttentiveMobileNetV3 | [search space](configs/_base_/nas_backbones/ofa_mobilenetv3_supernet.py) |    7.6    |  747.8   | 77.5  | [config](./detnas_subnet_frcnn_shufflenetv2_fpn_1x_coco.py) |                 [model](https://openmmlab-share.oss-cn-hangzhou.aliyuncs.com/mmrazor/v1/ofa/ofa_mobilenet_supernet_d234_e346_k357_w1_0.py_20221214_0940-d0ebc66f.pth)                  | Converted from the repo |
+| ImageNet | AttentiveMobileNetV3 |                   note8_lat@[email protected]_finetune@25                   |    4.3    |   70.9   | 70.3  |           [config](./OFA_SUBNET_NOTE8_LAT22.yaml)           | [model](https://openmmlab-share.oss-cn-hangzhou.aliyuncs.com/mmrazor/v1/ofa/ofa_mobilenet_subnet_8xb256_in1k_note8_lat%4022ms_top1%4070.4_finetune%4025.py_20221214_0938-fb7fb84f.pth) | Converted from the repo |
+| ImageNet | AttentiveMobileNetV3 |                   note8_lat@[email protected]_finetune@25                   |    4.6    |  105.4   | 72.6  |           [config](./OFA_SUBNET_NOTE8_LAT31.yaml)           | [model](https://openmmlab-share.oss-cn-hangzhou.aliyuncs.com/mmrazor/v1/ofa/ofa_mobilenet_subnet_8xb256_in1k_note8_lat%4031ms_top1%4072.8_finetune%4025.py_20221214_0939-981a8b2a.pth) | Converted from the repo |
+
+**Note**:
+
+1. OFA provides a more fine-grained search mode, which searches expand ratios & kernel size for each block in every layer of the defined supernet, therefore the subnet configs (format as .yaml) is more complex than those of BigNAS/AttentiveNAS.
+2. We product the [ofa script](../../../../tools/model_converters/convert_ofa_ckpt.py) to convert the official weight into MMRazor-style. The layer depth of a specific subnet is required when converting keys.
+3. The models above are converted from the [once-for-all official repo](https://github.com/mit-han-lab/once-for-all). The config files of these models
+   are only for inference. We don't ensure training accuracy of these config files and you are welcomed to contribute your reproduction results.
+
+## Citation
+
+```latex
+@inproceedings{
+  cai2020once,
+  title={Once for All: Train One Network and Specialize it for Efficient Deployment},
+  author={Han Cai and Chuang Gan and Tianzhe Wang and Zhekai Zhang and Song Han},
+  booktitle={International Conference on Learning Representations},
+  year={2020},
+  url={https://arxiv.org/pdf/1908.09791.pdf}
+}
+```