zhaowt61

No data

1

Can not find the data

I managed to make 77.24% by using step LR scheduler.

7

By using the following settings, I managed to achieve 77.24% at 24 epochs. Setting: 1 dataset: Pascal + SDB 2 4 opus, BatchSize = 16 3 Backbone: ResNet 4 —freeze-bn...

Can anyone reproduce the claimed accuracy?

1

I trained on Resnet using Pascal + SBD, but got an accuracy of 74.5%. All settings are default values.

Question about the training device on text2image

1

Thanks for the excellent work! I wonder about the training device on text2image. The paper says it is trained on a single A100, but it seems the settings in table...

One question about the brightness of the outcome from pertained model

4

Hi Dr. Wang, thanks for your excellent work! I noticed that the brightness of the outcome from the pertained model is significantly higher than the one I reproduced using exactly...

Question about the training schedule

4

Hi, I am wondering how many iterations does it take the achieve the showed results? I am using exactly the same training settings along with the dataset provided by the...

model doesn't converge

1

Hi, thanks for the excellent work. I tried to use the default settings and the datasets provided by the author, but the model doesn't converge (diverge after about 10 iterations)....

Hi team, thanks for the amazing work! I got the following error when I tried to finetune InternVL-Chat-1.5: [rank0]: Original Traceback (most recent call last): [rank0]: File "/opt/conda/envs/internvl_train2/lib/python3.9/site-packages/torch/nn/parallel/parallel_apply.py", line 83,...