topic Re: Deep Learning for land classification Error in ArcGIS Image Server Questions

Deep Learning for land classification Error

SherminHo — Sun, 14 Aug 2022 17:43:59 GMT

Dear Community,

attempted to classify pixels using deep learning by applying the Land Cover Classification dlpk.

However, an error returned: Error 999999

is there a way around this?

Thanks

Re: Deep Learning for land classification Error

SarahPC — Wed, 24 Mar 2021 01:57:45 GMT

I actually had the same error to following with interest.

Re: Deep Learning for land classification Error

Tim_McGinnes — Wed, 24 Mar 2021 02:54:57 GMT

Can you provide some more information to assist with diagnosing the problem -

- ArcGIS Pro version & how you installed the deep learning components

- Screenshot of the Messages in the geoprocessing output (it's down the bottom)

- Information on the input raster you are using - Type, Bands, Bits per Pixel, Columns, Rows, Coordinate System etc.

Re: Deep Learning for land classification Error

SherminHo — Wed, 24 Mar 2021 03:32:13 GMT

Hi Tim,

Thanks for responding.

I'm currently on ArcGIS Pro 2.7.2.

Installed the deep learning components by using the deep learning installer:

https://github.com/Esri/deep-learning-frameworks/blob/master/README.md?rmedium=links_esri_com_b_d&rsource=https%3A%2F%2Flinks.esri.com%2Fdeep-learning-framework-install

Once installed, I ran the model again but was unable to read the dlpk. I then attempted to reload the deep learning frameworks manually by following the steps in this link:- pls also see attached for the python cmd prompt i used

https://github.com/Esri/deep-learning-frameworks/blob/master/install-deep-learning-frameworks-manually-2-7.pdf

As for the information on the input raster, I was trying out with Multispectral Landsat imagery found on Living Atlas: https://landsat2.arcgis.com/arcgis/services/Landsat/MS/ImageServer

Re: Deep Learning for land classification Error

Tim_McGinnes — Wed, 24 Mar 2021 04:45:28 GMT

By default the Landsat service will show you a 3 band combination, but the deep learning model expects 7 bands. In the properties of the Landsat service under Processing Templates, you need to set the Processing Template to None. Worked for me ok once I did that. You should probably set your required extent in the environment variables as well.

Re: Deep Learning for land classification Error

SherminHo — Wed, 24 Mar 2021 07:13:03 GMT

Hi Tim,

Thanks for the tip. unfortunately, even after setting the processing template to None, and trying out on the extent in the example you have shared, i'm still receiving the same error message:

Re: Deep Learning for land classification Error

Tim_McGinnes — Wed, 24 Mar 2021 12:20:03 GMT

Well it's a slightly different error this time. My only suggestion now would be to go back to using the default Python environment (arcgispro-py3). What did you mean when you said it was "unable to read the dlpk"? I am also on Pro 2.7.2, and used the deep learning framework from the installer and using the default environment.

Re: Deep Learning for land classification Error

Anonymous User — Wed, 24 Mar 2021 16:27:57 GMT

Hi,

Yes you need to set the processing template to None.

Looking at the error It seems to me that you are trying to apply the sentinel 2 Landcover model on a Landsat 8 Imagery. I recommend you to confirm the model you are trying to use.

This model works with Landsat 8 Imagery https://www.arcgis.com/home/item.html?id=e732ee81a9c14c238a14df554a8e3225

and this model works with Sentinel 2 Imagery https://www.arcgis.com/home/item.html?id=afd124844ba84da69c2c533d4af10a58

Thanks,

Sandeep

Re: Deep Learning for land classification Error

SherminHo — Thu, 25 Mar 2021 09:45:22 GMT

Thanks for the advice Tim

Re: Deep Learning for land classification Error

SherminHo — Thu, 25 Mar 2021 09:46:32 GMT

Hi Sandeep,

Thanks for spotting the error. will rerun the model and update again 🙂

Re: Deep Learning for land classification Error

şevvalbulut — Wed, 13 Oct 2021 11:25:55 GMT

Hi ,

I have same error. Continuously giving an ERROR 999999. Runs the process 2% only then gives error

I'm currently on ArcGIS Pro 2.8.3

I use this model https://www.arcgis.com/home/item.html?id=e732ee81a9c14c238a14df554a8e3225 for landsat 8

I setted the processing template to None , i'm still receiving the same error message

Re: Deep Learning for land classification Error

Anonymous User — Wed, 13 Oct 2021 12:07:43 GMT

Hi,

Can you paste a screenshot of same ?

Thanks,

Sandeep

Re: Deep Learning for land classification Error

şevvalbulut — Wed, 13 Oct 2021 12:48:55 GMT

Hi Sandeep

Thanks

Re: Deep Learning for land classification Error

Anonymous User — Fri, 22 Oct 2021 04:30:31 GMT

Hi,

It seems that your training data had 4 or more bands but the image you are trying to deploy your model on has only 3 bands.

These models expect same number of bands in target imagery as in the training data.

You can try to retrain the model with fewer bands same as in target imagery, you might need to rexport the training data. If it was only alpha channel in the training imagery, you can just add one more constant band in the target imagery.

Thanks,

Sandeep

Re: Deep Learning for land classification Error

WillWoodroof — Wed, 27 Oct 2021 21:37:42 GMT

I'm having a similar issue when running the ArcGIS Pro Classification Wizard (Supervised/Object based/Support Vector Machine) on a local projected image clipped from Sentinel 2 Views. I get this error (screen shot attached) and can't proceed. Is this possible or am I missing something?

Thanks,

Will

Re: Deep Learning for land classification Error

GengshuoHu — Fri, 09 Sep 2022 02:00:05 GMT

hello

i also have a problem when using the deep learning

i am using the landsat 8 areial image and i download the deep learning as procedure.

Here is the message:

Classify Pixels Using Deep Learning
=====================
Parameters

Input Raster Surface Reflectance_LC08_L2SP_089079_20130427_20200912_02_T1_MTL
Output Classified Raster C:\Users\Admin\Documents\ArcGIS\Projects\MyProject1\MyProject1.gdb\SurfaceReflect_ClassifyPixel3
Model Definition C:\学习文件\2022 Semester 2\Final Research\土地利用深度学习\LandCoverClassification.dlpk
Arguments padding 128;batch_size 4;predict_background True;test_time_augmentation True;tile_size 512;landsat_imagery_level 2
Processing Mode PROCESS_AS_MOSAICKED_IMAGE
Output Folder
=====================
Environments

Extent 152.607318628634 -27.5730888984295 153.480255433056 -27.3123415412645 GEOGCS["GCS_GDA_1994",DATUM["D_GDA_1994",SPHEROID["GRS_1980",6378137.0,298.257222101]],PRIMEM["Greenwich",0.0],UNIT["Degree",0.0174532925199433]]
Cell Size 30
Processor Type GPU
=====================
Messages

Start Time: Friday, 9 September 2022 2:23:47 AM
ERROR 999999: Something unexpected caused the tool to fail. Contact Esri Technical Support (http://esriurl.com/support) to Report a Bug, and refer to the error help for potential solutions or workarounds.
Unable to read pixels from the python raster function.
Unable to read pixels from the python raster function.
Function Read Error [Raster Function Template]
Unable to read pixels from the python raster function.
Traceback (most recent call last):
File "C:\Users\Admin\AppData\Local\Temp\ArcGISProTemp14780\LandCoverClassification.dlpk\ArcGISImageClassifier.py", line 275, in updatePixels
xx = self.child_image_classifier.updatePixelsTTA(tlc, shape, props, **pixelBlocks).astype(props['pixelType'], copy=False)
File "C:\Users\Admin\AppData\Local\Temp\ArcGISProTemp14780\LandCoverClassification.dlpk\_unet.py", line 443, in updatePixelsTTA
all_activations = self.tta_predict(
File "C:\Users\Admin\AppData\Local\Temp\ArcGISProTemp14780\LandCoverClassification.dlpk\_unet.py", line 399, in tta_predict
int_surface = self.split_predict_interpolate(
File "C:\Users\Admin\AppData\Local\Temp\ArcGISProTemp14780\LandCoverClassification.dlpk\_unet.py", line 371, in split_predict_interpolate
output = self.model(patches)
File "C:\Program Files\ArcGIS\Pro\bin\Python\envs\arcgispro-py3\Lib\site-packages\torch\nn\modules\module.py", line 889, in _call_impl
result = self.forward(*input, **kwargs)
File "C:\Program Files\ArcGIS\Pro\bin\Python\envs\arcgispro-py3\Lib\site-packages\fastai\layers.py", line 136, in forward
nres = l(res)
File "C:\Program Files\ArcGIS\Pro\bin\Python\envs\arcgispro-py3\Lib\site-packages\torch\nn\modules\module.py", line 889, in _call_impl
result = self.forward(*input, **kwargs)
File "C:\Program Files\ArcGIS\Pro\bin\Python\envs\arcgispro-py3\Lib\site-packages\fastai\layers.py", line 136, in forward
nres = l(res)
File "C:\Program Files\ArcGIS\Pro\bin\Python\envs\arcgispro-py3\Lib\site-packages\torch\nn\modules\module.py", line 889, in _call_impl
result = self.forward(*input, **kwargs)
File "C:\Program Files\ArcGIS\Pro\bin\Python\envs\arcgispro-py3\Lib\site-packages\fastai\layers.py", line 150, in forward
def forward(self, x): return torch.cat([x,x.orig], dim=1) if self.dense else (x+x.orig)
RuntimeError: CUDA out of memory. Tried to allocate 412.00 MiB (GPU 0; 2.00 GiB total capacity; 1.11 GiB already allocated; 0 bytes free; 1.41 GiB reserved in total by PyTorch)
Python raster function's .updatePixels() method returned nothing.
Unable to read pixels from the python raster function.
Unable to read pixels from the python raster function.
Function Read Error [Raster Function Template]
Unable to read pixels from the python raster function.
Traceback (most recent call last):
File "C:\Users\Admin\AppData\Local\Temp\ArcGISProTemp14780\LandCoverClassification.dlpk\ArcGISImageClassifier.py", line 275, in updatePixels
xx = self.child_image_classifier.updatePixelsTTA(tlc, shape, props, **pixelBlocks).astype(props['pixelType'], copy=False)
File "C:\Users\Admin\AppData\Local\Temp\ArcGISProTemp14780\LandCoverClassification.dlpk\_unet.py", line 443, in updatePixelsTTA
all_activations = self.tta_predict(
File "C:\Users\Admin\AppData\Local\Temp\ArcGISProTemp14780\LandCoverClassification.dlpk\_unet.py", line 399, in tta_predict
int_surface = self.split_predict_interpolate(
File "C:\Users\Admin\AppData\Local\Temp\ArcGISProTemp14780\LandCoverClassification.dlpk\_unet.py", line 371, in split_predict_interpolate
output = self.model(patches)
File "C:\Program Files\ArcGIS\Pro\bin\Python\envs\arcgispro-py3\Lib\site-packages\torch\nn\modules\module.py", line 889, in _call_impl
result = self.forward(*input, **kwargs)
File "C:\Program Files\ArcGIS\Pro\bin\Python\envs\arcgispro-py3\Lib\site-packages\fastai\layers.py", line 136, in forward
nres = l(res)
File "C:\Program Files\ArcGIS\Pro\bin\Python\envs\arcgispro-py3\Lib\site-packages\torch\nn\modules\module.py", line 889, in _call_impl
result = self.forward(*input, **kwargs)
File "C:\Program Files\ArcGIS\Pro\bin\Python\envs\arcgispro-py3\Lib\site-packages\fastai\layers.py", line 136, in forward
nres = l(res)
File "C:\Program Files\ArcGIS\Pro\bin\Python\envs\arcgispro-py3\Lib\site-packages\torch\nn\modules\module.py", line 889, in _call_impl
result = self.forward(*input, **kwargs)
File "C:\Program Files\ArcGIS\Pro\bin\Python\envs\arcgispro-py3\Lib\site-packages\fastai\layers.py", line 150, in forward
def forward(self, x): return torch.cat([x,x.orig], dim=1) if self.dense else (x+x.orig)
RuntimeError: CUDA out of memory. Tried to allocate 412.00 MiB (GPU 0; 2.00 GiB total capacity; 1.11 GiB already allocated; 0 bytes free; 1.41 GiB reserved in total by PyTorch)
Python raster function's .updatePixels() method returned nothing.
Unable to read pixels from the python raster function.
Unable to read pixels from the python raster function.
Function Read Error [Raster Function Template]
Unable to read pixels from the python raster function.
Traceback (most recent call last):
File "C:\Users\Admin\AppData\Local\Temp\ArcGISProTemp14780\LandCoverClassification.dlpk\ArcGISImageClassifier.py", line 275, in updatePixels
xx = self.child_image_classifier.updatePixelsTTA(tlc, shape, props, **pixelBlocks).astype(props['pixelType'], copy=False)
File "C:\Users\Admin\AppData\Local\Temp\ArcGISProTemp14780\LandCoverClassification.dlpk\_unet.py", line 443, in updatePixelsTTA
all_activations = self.tta_predict(
File "C:\Users\Admin\AppData\Local\Temp\ArcGISProTemp14780\LandCoverClassification.dlpk\_unet.py", line 399, in tta_predict
int_surface = self.split_predict_interpolate(
File "C:\Users\Admin\AppData\Local\Temp\ArcGISProTemp14780\LandCoverClassification.dlpk\_unet.py", line 371, in split_predict_interpolate
output = self.model(patches)
File "C:\Program Files\ArcGIS\Pro\bin\Python\envs\arcgispro-py3\Lib\site-packages\torch\nn\modules\module.py", line 889, in _call_impl
result = self.forward(*input, **kwargs)
File "C:\Program Files\ArcGIS\Pro\bin\Python\envs\arcgispro-py3\Lib\site-packages\fastai\layers.py", line 136, in forward
nres = l(res)
File "C:\Program Files\ArcGIS\Pro\bin\Python\envs\arcgispro-py3\Lib\site-packages\torch\nn\modules\module.py", line 889, in _call_impl
result = self.forward(*input, **kwargs)
File "C:\Program Files\ArcGIS\Pro\bin\Python\envs\arcgispro-py3\Lib\site-packages\fastai\layers.py", line 136, in forward
nres = l(res)
File "C:\Program Files\ArcGIS\Pro\bin\Python\envs\arcgispro-py3\Lib\site-packages\torch\nn\modules\module.py", line 889, in _call_impl
result = self.forward(*input, **kwargs)
File "C:\Program Files\ArcGIS\Pro\bin\Python\envs\arcgispro-py3\Lib\site-packages\fastai\layers.py", line 150, in forward
def forward(self, x): return torch.cat([x,x.orig], dim=1) if self.dense else (x+x.orig)
RuntimeError: CUDA out of memory. Tried to allocate 412.00 MiB (GPU 0; 2.00 GiB total capacity; 1.11 GiB already allocated; 0 bytes free; 1.41 GiB reserved in total by PyTorch)
Python raster function's .updatePixels() method returned nothing.
Failed to execute (ClassifyPixelsUsingDeepLearning).
Failed at Friday, 9 September 2022 2:25:08 AM (Elapsed Time: 1 minutes 20 seconds)

Re: Deep Learning for land classification Error

Nariman77 — Tue, 23 May 2023 10:34:23 GMT

I also have error 9999, but I used Sentinel-2. And here I used 3 bands