search

AlibabaResearch / efficientteacher

A Supervised and Semi-Supervised Object Detection Library for YOLO Series
GNU General Public License v3.0
820 stars 147 forks source link

Unlabelled data Training #80

Open jaideep11061982 opened 1 year ago

jaideep11061982 commented 1 year ago

Thanks for wonder for repo train SSOD model from scratch, the default setting is to first conduct a 220-epoch burn-in training, and then enter the SSOD training, which has been introduced in our paper.

Could you help understand the above line @alibaba-oss

1_ Also brief steps to follow to label unlabelled data. 2) Which model to use to label unlabelled data at the end of training

3) What is purpose of "https://github.com/AlibabaResearch/efficientteacher/blob/main/configs/ssod/custom/yolov5l_transfer_ssod.yaml" which one to use for custom data

4) IS the labelled data is fraction of total data which we label and rest we label through the efficient teacher method

5) modify the train: data/custom_train.txt in yolov5l_custom.yaml, and then enter the following script. export CUDA_VISIBLE_DEVICES="0,1,2,3,4,5,6,7" python -m torch.distributed.launch --nproc_per_node 8 --master_addr 127.0.0.2 --master_port 29502 train

Supervised training step is to train on how much amount of labelled data ? what to modify in this step of supervised ?

jaideep11061982 commented 1 year ago

@alibaba-oss @meton-robean @kdy1999