I have read a Dataset in NC file using netCDF4 in python and then tried to plot the dataset using matplotlib.pyplot. But issue is the plot shows the image but with 90 Degrees clockwise rotation. Following code I have used: from netCDF4 import Dateset import matplotlib.pyplot as plt f = Dataset(r'F:\data\data.nc', 'r') data = f['et'][:] plt.imshow(data) plt.show() One doubt I am having, I’d that I have 2 NC files having same dataset and in that, dataset in one file is getting properly displayed by using np.rot90(data) which was suggested to me in the answer given below and in another file I need not to do "np.rot90(data)" but instead I need to do "data[::-1]" to display the data correctly and interesting thing is when I am importing these 2 files directly in the ArcMap both are getting properly displayed. Is the issue with NC data?

I'm new on GEE and I'm trying to perform an accuracy assessment of a supervised classification from S1 of a flood area, by the use of a confusion matrix. Once the model is created, in order to evaluate it, I imported a shapefile with data relating the reality on the ground regarding the flood, and from this shapefile I selected a sample of polygons of flooded and non-flooded areas, created a new feature collection and then through the confusion matrix I wanted to compare this new dateset with the model to obtain an evaluation of the classification. This is the code I used: //create a feature collection of samples (samples are polygons set as geometry). var newFc = ee.FeatureCollection( [ee.Feature(water_permanent, {'landCover':1} ), ee.Feature(water_flooded, {'landCover':2} ), ee.Feature(urban, {'landCover':3} ), ee.Feature(agricultural_field,{'landCover’:4})] ); //Overlay the points on the imagery to get training //(UnionMed is the starting image from where I took the samples). var training = unionMed.sampleRegions({ collection: newFc, scale:10, properties:['landCover'], tileScale:16 }); //Define the classifier and train it. var classifier = ee.Classifier.smileRandomForest(100).train(training, "landCover"); //Classify the image. var classified = unionMed.classify(classifier); //Accuracy Asssesment. //define a new FC from the reference shapefile. var newFc_test = ee.FeatureCollection( [ee.Feature(No_WF_test, {'landCover_test':1} ), ee.Feature(WF_test, {'landCover_test’:2})]); var validated = classified.sampleRegions({ collection: newFc_test, properties: ['landCover_test'], scale:10, tileScale:16 }); //Classify the validation data. var validation=validatio.classify(classifier); //Get a confusion matrix representing resubstitution accuracy. var testAccuracy = validation.errorMatrix('landCover','landCover_test') print('Confusion matrix:', testAccuracy) print('Accuracy:', testAccuracy.accuracy()) The problem is that I get this error: ConfusionMatrix (Error) Collection.errorMatrix: Property 'VV' of feature '0_0' is missing.

Advanced Visualization Application

Signed PaperbackOctober 20, 2026 Publication DateSet in post-WWII Paris, the wife of an American foreign correspondent stumbles on the body of her murde...

Signed PaperbackOctober 20, 2026 Publication DateSet in post-WWII Paris, the wife of an American foreign correspondent stumbles on the body of her murde...

Signed PaperbackOctober 20, 2026 Publication DateSet in post-WWII Paris, the wife of an American foreign correspondent stumbles on the body of her murde...

I have read a Dataset in NC file using netCDF4 in python and then tried to plot the dataset using matplotlib.pyplot. But issue is the plot shows the image but with 90 Degrees clockwise rotation. Following code I have used: from netCDF4 import Dateset import matplotlib.pyplot as plt f = Dataset(r'F:\data\data.nc', 'r') data = f['et'][:] plt.imshow(data) plt.show() One doubt I am having, I’d that I have 2 NC files having same dataset and in that, dataset in one file is getting properly displayed by using np.rot90(data) which was suggested to me in the answer given below and in another file I need not to do "np.rot90(data)" but instead I need to do "data[::-1]" to display the data correctly and interesting thing is when I am importing these 2 files directly in the ArcMap both are getting properly displayed. Is the issue with NC data?

I'm new on GEE and I'm trying to perform an accuracy assessment of a supervised classification from S1 of a flood area, by the use of a confusion matrix. Once the model is created, in order to evaluate it, I imported a shapefile with data relating the reality on the ground regarding the flood, and from this shapefile I selected a sample of polygons of flooded and non-flooded areas, created a new feature collection and then through the confusion matrix I wanted to compare this new dateset with the model to obtain an evaluation of the classification. This is the code I used: //create a feature collection of samples (samples are polygons set as geometry). var newFc = ee.FeatureCollection( [ee.Feature(water_permanent, {'landCover':1} ), ee.Feature(water_flooded, {'landCover':2} ), ee.Feature(urban, {'landCover':3} ), ee.Feature(agricultural_field,{'landCover’:4})] ); //Overlay the points on the imagery to get training //(UnionMed is the starting image from where I took the samples). var training = unionMed.sampleRegions({ collection: newFc, scale:10, properties:['landCover'], tileScale:16 }); //Define the classifier and train it. var classifier = ee.Classifier.smileRandomForest(100).train(training, "landCover"); //Classify the image. var classified = unionMed.classify(classifier); //Accuracy Asssesment. //define a new FC from the reference shapefile. var newFc_test = ee.FeatureCollection( [ee.Feature(No_WF_test, {'landCover_test':1} ), ee.Feature(WF_test, {'landCover_test’:2})]); var validated = classified.sampleRegions({ collection: newFc_test, properties: ['landCover_test'], scale:10, tileScale:16 }); //Classify the validation data. var validation=validatio.classify(classifier); //Get a confusion matrix representing resubstitution accuracy. var testAccuracy = validation.errorMatrix('landCover','landCover_test') print('Confusion matrix:', testAccuracy) print('Accuracy:', testAccuracy.accuracy()) The problem is that I get this error: ConfusionMatrix (Error) Collection.errorMatrix: Property 'VV' of feature '0_0' is missing.

I have read a Dataset in NC file using netCDF4 in python and then tried to plot the dataset using matplotlib.pyplot. But issue is the plot shows the image but with 90 Degrees clockwise rotation. Following code I have used: from netCDF4 import Dateset import matplotlib.pyplot as plt f = Dataset(r'F:\data\data.nc', 'r') data = f['et'][:] plt.imshow(data) plt.show() One doubt I am having, I’d that I have 2 NC files having same dataset and in that, dataset in one file is getting properly displayed by using np.rot90(data) which was suggested to me in the answer given below and in another file I need not to do "np.rot90(data)" but instead I need to do "data[::-1]" to display the data correctly and interesting thing is when I am importing these 2 files directly in the ArcMap both are getting properly displayed. Is the issue with NC data?

I'm new on GEE and I'm trying to perform an accuracy assessment of a supervised classification from S1 of a flood area, by the use of a confusion matrix. Once the model is created, in order to evaluate it, I imported a shapefile with data relating the reality on the ground regarding the flood, and from this shapefile I selected a sample of polygons of flooded and non-flooded areas, created a new feature collection and then through the confusion matrix I wanted to compare this new dateset with the model to obtain an evaluation of the classification. This is the code I used: //create a feature collection of samples (samples are polygons set as geometry). var newFc = ee.FeatureCollection( [ee.Feature(water_permanent, {'landCover':1} ), ee.Feature(water_flooded, {'landCover':2} ), ee.Feature(urban, {'landCover':3} ), ee.Feature(agricultural_field,{'landCover’:4})] ); //Overlay the points on the imagery to get training //(UnionMed is the starting image from where I took the samples). var training = unionMed.sampleRegions({ collection: newFc, scale:10, properties:['landCover'], tileScale:16 }); //Define the classifier and train it. var classifier = ee.Classifier.smileRandomForest(100).train(training, "landCover"); //Classify the image. var classified = unionMed.classify(classifier); //Accuracy Asssesment. //define a new FC from the reference shapefile. var newFc_test = ee.FeatureCollection( [ee.Feature(No_WF_test, {'landCover_test':1} ), ee.Feature(WF_test, {'landCover_test’:2})]); var validated = classified.sampleRegions({ collection: newFc_test, properties: ['landCover_test'], scale:10, tileScale:16 }); //Classify the validation data. var validation=validatio.classify(classifier); //Get a confusion matrix representing resubstitution accuracy. var testAccuracy = validation.errorMatrix('landCover','landCover_test') print('Confusion matrix:', testAccuracy) print('Accuracy:', testAccuracy.accuracy()) The problem is that I get this error: ConfusionMatrix (Error) Collection.errorMatrix: Property 'VV' of feature '0_0' is missing.