GitHub - JuliusFx131/GeoAI-Estimating-Soil-Parameters-Using-Hyperspectral-Images: The traditional in-situ soil analysis methods are laborious & inefficient, limiting scalability and hindering timely access to crucial soil data for optimal fertilization by farmers. In the amazing challenge, we tried to predict soil parameters(Phosphorous, Potassium, Magnesium and pH)from hyperspectral satellite images.

JuliusFx131 / GeoAI-Estimating-Soil-Parameters-Using-Hyperspectral-Images Public

forked from ITU-GeoAI-Challenge/Estimating-Soil-Parameters-from-Hyperspectral-Images-by-ITU-2023

Notifications You must be signed in to change notification settings
Fork 0
Star 2

The traditional in-situ soil analysis methods are laborious & inefficient, limiting scalability and hindering timely access to crucial soil data for optimal fertilization by farmers. In the amazing challenge, we tried to predict soil parameters(Phosphorous, Potassium, Magnesium and pH)from hyperspectral satellite images.

zindi.africa/competitions/geoai-challenge-estimating-soil-parameters-from-hyperspectral-images

2 stars 1 fork Branches Tags Activity

Star

Notifications

Branches Tags

Name		Name	Last commit message	Last commit date
Latest commit History 4 Commits
1_BASELINES_Final-HIST.ipynb		1_BASELINES_Final-HIST.ipynb
1_BASELINES_Final-LGBM.ipynb		1_BASELINES_Final-LGBM.ipynb
Data Prep.ipynb		Data Prep.ipynb
ENSEMBLE.ipynb		ENSEMBLE.ipynb
HIST.csv		HIST.csv
LGBM.csv		LGBM.csv
LGB_REDUCED_50_-HIST_100_.csv		LGB_REDUCED_50_-HIST_100_.csv
README.txt		README.txt
Technical Report.pdf		Technical Report.pdf
presentation.pptx		presentation.pptx
requirements.txt		requirements.txt

Repository files navigation

1. The environement used was anaconda on my local 16gb ram laptop

2. All files should reside in one folder

3. Run the codes in the following manner:

	a) 1_BASELINES_Final-HIST.ipynb (approx 5 minutes)-this generates "HIST.csv" file
	b) 1_BASELINES_Final-LGBM.ipynb (approx 3 minutes)-this generates "LGBM.csv" file
	c) ENSEMBLE.ipynb (approx 1 minute)-this generates "LGB_REDUCED_50_-HIST_100_.csv" file

4. This should be able to generate

NB:
	If you would wish to generate the "test_df.csv" and "train_df.csv" used in this work, you may wish to run the "Data Prep.ipynb" 
	This was run on free kaggle GPUP100 because of the challenges the free google colab gpu posed.