developmentseed / satTS

ML pipeline to classify crop types with multi-spectral and multi-temporal EO data
MIT License
18 stars 8 forks source link

Blog post 1 #2

Open JmeCS opened 6 years ago

JmeCS commented 6 years ago

Please let me know if a Github Issue is not the correct way to track this.

Link to draft of blog post 1 for this project: Draft

I'm not sure how well Medium tracks changes but you should be able to leave notes and make edits.

Please confirm once you've it looked over (if you want to). If you do not plan on making edits, let me know. I'll publish once all feedback is received and incorporated.

@ianschuler @abarciauskas-bgse @matthewhanson

abarciauskas-bgse commented 6 years ago

@JmeCS awesome! Thanks for putting all this together! I made some minor comments but higher level ones would be:

so perhaps:

In partnership with the Earth on AWS fellowship program and using the Earth on AWS platform, I am working with Development Seed to explore possible approaches to overcoming problems inherent to crop classification in developing countries.

(that could still use some tweaking, admittedly)

JmeCS commented 6 years ago

@ianschuler @abarciauskas-bgse @matthewhanson

Thanks very much for the edits and comments.

I've completed a second draft with a much less technical and focused introductory section. Link:

https://medium.com/@jamey_20161/towards-a-global-view-of-within-season-crop-classification-4eb4c4437ffe

I'm still not especially happy with the title. Any suggestions there would be appreciated. I'll get it published once I have sign-off from @ianschuler

ianschuler commented 6 years ago

Predicting crop estimates worldwide A novel deep learning approach to predict crop types from open satellite imagery.

^ still needs some love, but maybe better? Don't let me bottleneck this. Look for Calliope in @dereklieu @drewbo @matthewhanson