Share parameters

[deep404]

Hi guys, Thank you for sharing your amazing work! For training I am using the dataset posted here https://app.activeloop.ai/crossvivit/SunLake, which can be accessed with deeplake. Furthermore, I am using some of the parameters which you have shared in your paper, however, I got some errors that mat1 could not be multiplied with mat2, because of different matrix size. In order to solve this little problem, I have changed ctx_channels from 8 to 23, ts_channels from 16 to 8 in cross_vivit_bis.yaml. However, I am not feeling that it is the correct solution to the problem. Could you please share all your parameters from cross_vivit_bis.yaml, txcontext_datamodule.yaml and cross_vivit.yaml?

[yy0127-ai]

Hello friend, I would like to ask how to download a dataset and run the program? I encountered this error ，deeplake.util.exceptions.DatasetHandlerError: A Deep Lake dataset does not exist at the given path (/network/scratch/g/ghait.boukachab/o/EUMETSAT/tsf_eumetsat_bis-64_64). Check the path provided or in case you want to create a new dataset, use deeplake.empty(). Perhaps this is a very simple question, but it has troubled me for a long time. Looking forward to your answer, thank you very much.

[deep404]

I have added the parameter reset = True -> self.deeplake_ds = deeplake.load(self.data_dir, reset=True). It has solved the problem for me. I used the command python main.py to start training, but I am not sure, it is correct

[yy0127-ai]

Thanks for your answer. Sorry to ask such a simple question. Mainly, I don't know how to download the dataset from Deep Lake.

[deep404]

It is not necessary to download the dataset from deeplake. Just go to configs/paths/default.yaml and change data_dir param. Put there the link to deeplake dataset: "data_dir: hub://crossvivit/SunLake"

[yy0127-ai]

Thank you very much!

[jaggbow]

Thank you @deep404 for helping @yy0127-ai with the issue. We're going to update the configs accordingly. Please note that the parameters used in the papers will be the ones assigned in configs/experiment.

[deep404]

My please to help other. Great, I'm looking forward to it

[jaggbow]

Hey,

I want to add that for any model you run, you have to specify experiment=model_you_want to the command so that it knows which parameters to run, this way you won't have any problems. Can you try that and tell me if the problem persists?

[liujian123223]

My please to help other. Great, I'm looking forward to it

Hello friend, I would like to ask how to download a dataset and run the program? I encountered this error ，deeplake.util.exceptions.DatasetHandlerError: A Deep Lake dataset does not exist at the given path (/network/scratch/g/ghait.boukachab/o/EUMETSAT/tsf_eumetsat_bis-64_64). Check the path provided or in case you want to create a new dataset, use deeplake.empty(). Perhaps this is a very simple question, but it has troubled me for a long time. Looking forward to your answer, thank you very much.

Hello, do you know how to run crosslive on Linux? I only have one computer and haven't used Slurm before. I don't know how to run it. I would greatly appreciate it if you could provide assistance

[liujian123223]

It is not necessary to download the dataset from deeplake. Just go to configs/paths/default.yaml and change data_dir param. Put there the link to deeplake dataset: "data_dir: hub://crossvivit/SunLake"

Hi, I am importing sunlake dataset by using deeplake but the process is taking a long time (more than 2 hours) which is causing me to get stuck here without outputting the results. May I ask if you encounter this problem?

Is it possible to download the Sunlake dataset and use it locally? Looking forward to your reply!

import deeplake

ds = deeplake.load('hub://crossvivit/SunLake')

[liujian123223]

Thank you very much!

"Excuse me, have you successfully imported the SunLake dataset from Deeplake? I'm having a lot of trouble because it's very slow when I use ds = deeplake.load('hub://crossvivit/SunLake'). Can I discuss this issue with you?"