mirrors/mmselfsup
1
0
Fork 0
mirror of https://github.com/open-mmlab/mmselfsup.git synced 2025-06-03 14:59:38 +08:00
Code Issues Packages Projects Releases Wiki Activity

Update GETTING_STARTED.md

Browse Source
This commit is contained in:
Xiaohang Zhan 2020-06-16 00:57:17 +08:00 committed by GitHub
parent 8ec52fe68f
commit bcca9af72d
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 10 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

20
docs/GETTING_STARTED.md
View File

@ -84,15 +84,15 @@ Augments:
- `WEIGHT_FILE`: the output backbone weights file, e.g., `odc_v1.pth`.
2. Next, run detection. For more details to setup the environments for detection, please refer [here](benchmarks/detection/README.md).
```shell
conda activate detectron2
cd benchmarks/detection
python convert-pretrain-to-detectron2.py ${WEIGHT_FILE} ${OUTPUT_FILE} # must use .pkl as the output extension.
bash run.sh ${DET_CFG} ${OUTPUT_FILE}
```
Arguments:
- `DET_CFG`: the detectron2 config file, usually we use `configs/pascal_voc_R_50_C4_24k_moco.yaml`.
- `OUTPUT_FILE`: converted backbone weights file, e.g., `odc_v1.pkl`.
```shell
conda activate detectron2
cd benchmarks/detection
python convert-pretrain-to-detectron2.py ${WEIGHT_FILE} ${OUTPUT_FILE} # must use .pkl as the output extension.
bash run.sh ${DET_CFG} ${OUTPUT_FILE}
```
Arguments:
- `DET_CFG`: the detectron2 config file, usually we use `configs/pascal_voc_R_50_C4_24k_moco.yaml`.
- `OUTPUT_FILE`: converted backbone weights file, e.g., `odc_v1.pkl`.
**Note**:
- This benchmark must use 8 GPUs as the default setting from MoCo.
@ -136,7 +136,7 @@ python tools/publish_model.py ${WEIGHT_FILE}
You may refer to existing modules under respective folders.
#### Features may facilitate your implementation
#### Features that may facilitate your implementation
* Decoupled data source and dataset.