New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
fail to reproduce #2
Comments
Have you trained the network for four rounds? And each round should load the model trained on the previous round excepted that for the first round, ImageNet-pretrained should be loaded. |
I have done the 3rd stage and used the weights from the previous stage for sure. Yet, the results of the 3rd stage remained poor (mAP_E < 70, mAP_M < 50 and mAP_H < 20 on ROxford). I am training the 4th stage now, with no clear improvement at the moment. |
Hi, I just finished stage 4 for GLDv2, with resnet101+GeM(p=3)+Head(norm_linear_norm) as given in the configs. The reproduced results are here:
The results taken from the Table 3 are much better, so what maybe wrong ?
|
Sorry for the late replay. I believe the problem should be caused by the first three rounds of training (base stages). You can download the trained model (after three rounds of training) and start the 4th training (use the impr config), then you should reach the performance reported in the paper. Checkpoint can be downloaded here. Note that you may need to change some parameter name to load the checkpoint. |
Hi, thanks for the checkpoint, and I am going to have a try. I notice that this checkpoint contains keys like "combiner.spatial_attention.conv1.weight", which belong to the spatial attention module. In this git repository, the spatial_attention is off in both config and forward function. Does it help in your experiments? |
Hi, I have done the 4th stage with your ckpt, and the results are similar to the ones on paper.
How do you get this ckpt of the 3rd stage? Anything I missed? Thanks! |
Thanks for sharing your code. @zeludeng I am also reproducing with your code (all stages..). Also, the evaluation using the trained model (s3_pretrained.pth) you shared above is very low. CUDA_VISIBLE_DEVICES=2 python tools/test_net.py --cfg configs/gldv2/impr.yaml
|
Hi, @peternara ! I think I can answer your 2nd question. But I also failed to reproduce from scratch myself, and the reason is unclear.
|
Ooh I didn't read carefully. However, I can't understand why it can't be reproduced with this published code. |
Hi, I am reproducing the work. Following the readme, I train the resnet101 with default settings except for a little path changes.
However, the evaluation results on ROxford(no 1M distractors) is too poor, and I get mAP_E < 70, mAP_M < 50 and mAP_H < 20。Similar things on the RParis evaluation.
The text was updated successfully, but these errors were encountered: