-
-
Notifications
You must be signed in to change notification settings - Fork 246
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
Classifier sand color option #176
Comments
hi kat, |
Hi Kilian,
Thanks for getting back to me.
I am currently doing precisely that: retraining the classifier and assessing the resulting models. My area of interest is the SW of England: lots of very dissipative beaches! Even though I would not say the sediment is not very light in colour, the "bright" classifier performs better. What you say makes sense, so now I have a couple more questions 😎:
1. Would I be right in assuming that the training dataset that is available in zenodo and figshare are the ones used to train the "default" versions (trained on Collaroy, Byron, Newcastle, Sawtel and Motuya), and the "bright" and "dark" models were trained on different beaches?
2. What made you decide to train different classifiers for Landsat but not S2?
3. What made you decide to train the Landsat classifier on L8 only?
4. (last one, I promise)... When I get to evaluating the new classifier, I get an "invalid input type" error (see attached pdf). Again, this is a low priority issue but if the problem is something obvious, it would be nice to sort it out.
I'd like to thank you for all your help and for making CoastSat available in the first place! The number of users it already has, is a testament to how useful a tool it is!
Cheers,
Kat.
…________________________________
From: Kilian Vos <[email protected]>
Sent: Saturday, February 6, 2021 3:18:49 AM
To: kvos/CoastSat <[email protected]>
Cc: Aikaterini Konstantinou <[email protected]>; Author <[email protected]>
Subject: Re: [kvos/CoastSat] Classifier sand color option (#176)
hi kat,
I trained the classifier with different training datasets, for example the 'bright' sand, I digitsed pixels on beaches with extremely white sand. You can create training data and re-train de classifier following the instructions here https://github.com/kvos/CoastSat/blob/master/doc/train_new_classifier.md<https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fkvos%2FCoastSat%2Fblob%2Fmaster%2Fdoc%2Ftrain_new_classifier.md&data=04%7C01%7Ca.konstantinou%40plymouth.ac.uk%7Cb0346570112345d70d6408d8ca4decd7%7C5437e7eb83fb4d1abfd3bb247e061bf1%7C1%7C0%7C637481783323112672%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=NwLilIiZD8H2Odj2QY0Xw2NVtvrGMwqVtJmLklm1G24%3D&reserved=0>
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub<https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fkvos%2FCoastSat%2Fissues%2F176%23issuecomment-774388782&data=04%7C01%7Ca.konstantinou%40plymouth.ac.uk%7Cb0346570112345d70d6408d8ca4decd7%7C5437e7eb83fb4d1abfd3bb247e061bf1%7C1%7C0%7C637481783323122667%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=AS94yQFoQJE%2FeyFsc15hZWgzmxb0xWimWLsKbACqibE%3D&reserved=0>, or unsubscribe<https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fnotifications%2Funsubscribe-auth%2FAMFFAT6OF2GUFCTPSMFZPS3S5SYJTANCNFSM4XCZHRDQ&data=04%7C01%7Ca.konstantinou%40plymouth.ac.uk%7Cb0346570112345d70d6408d8ca4decd7%7C5437e7eb83fb4d1abfd3bb247e061bf1%7C1%7C0%7C637481783323132659%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=F%2FodQgnkTUNmB2VzoCcypDwIeOGI3eIW%2BmDZkq%2FmwF8%3D&reserved=0>.
________________________________
[http://www.plymouth.ac.uk/images/email_footer.gif]<http://www.plymouth.ac.uk/worldclass>
This email and any files with it are confidential and intended solely for the use of the recipient to whom it is addressed. If you are not the intended recipient then copying, distribution or other use of the information contained is strictly prohibited and you should not rely on it. If you have received this email in error please let the sender know immediately and delete it from your system(s). Internet emails are not necessarily secure. While we take every care, University of Plymouth accepts no responsibility for viruses and it is your responsibility to scan emails and their attachments. University of Plymouth does not accept responsibility for any changes made after it was sent. Nothing in this email or its attachments constitutes an order for goods or services unless accompanied by an official order form.
|
hi Kat,
thanks, also it would be great if you could have a look at this #177 and maybe provide some of your observations as well |
Hi Kilian,
-Kat |
|
problem 4. (evaluating the classifier) is fixed in the latest update #192 |
Hi Kilian,
I am testing different classifier models and am curious as to how you created the classifier models for the different sand colour options. I am guessing the "default" options is the one derived directly from training the classifier as described in your papers, but I'm not quite sure how the other two models were derived.
This is a low priority question but it would be great if you found the time to get it...
Cheers
Kat
The text was updated successfully, but these errors were encountered: