Latest Contributions by TimGrenside

‎03-10-2021

Hmmm, this is mine (image below). The left side 32bit unsigned, right side 8bit unsigned. Other thoughts - Your 'Map' set to WGS 1984 Web Mercator? - Cell size set to 10 on the Classify Pixels for Deep Learning?

‎03-08-2021

I got strange results but found the following helped: -Change the processing template to None of the original Sentinel View (as mentioned) -Right-Click on the Sentinel View layer -> Data -> Export Raster -Set the Clipping Geometry to "Current Display Extent" (as a test) -Create the TIFF with these settings, Pixel Type to 32 bit unsigned and Output Format to TIFF The nvidia-smi.exe output

‎03-02-2021

Hi Angus, Thanks for your response. I added the registry setting and environment variable CUDA_VISIBLE_DEVICES, but it doesn't seem to make a difference and the issue is still happening. Regards Tim

‎02-26-2021

Hi Sandeep Thanks for your response. If I change the batch size to 1 and run the python code it still checkerboards. It still feels like it is running out of memory, is there a way to see any python/ArcGIS logs? Regards Tim

‎02-21-2021

Hi Sandeep Yes - once the process checkerboards after a python run - the 'History' (or original Geoprocess tool) will not work, until I close and restart ArcGIS Pro. So it is like once it checkboards the only way to get it to work again is too restart ArcGIS Pro. Many thanks Tim

‎02-11-2021

Wait what, yes I am using that. I thought that was the latest as it was updated with ArcGIS pro 2.7 What should I be using instead? Many thanks

‎02-10-2021

Im running ArcGIS 2.7.1 and having issues with Classify Pixels Using Deep Learning. It runs fine when running it through the ArcGIS Pro greprocessing tool. But if I go to the History of the job just run and click “Send to Python Window”, the python command checkboards the output raster (below image). I think I have tracked it down to possible GPU memory issues. The ArcGIS geoprocess and python must be using different code, as from the Task Manager, the GPU memory signature is completely different and in fact the python code doesn’t release the Dedicate GPU memory when its finished (see below). Additional information: - The model is running resnet50, unet with fastai (resnet34 doesn’t seem to have this issue and works fine in python) -The GPU is a GeForce RTX 2080 Ti (so 11 GBs) - Changing the batch size to 1 doesn’t make a difference - I think the python code might be crashing, but not reporting any issues to ArcGIS. Has anyone managed to get similar to above working with resnet50 in python? Does anybody have any ideas on what I could try to get it working? Many Thanks Tim GPU task manager when running through ArcGIS geoprocess tool GPU task manager when running through python window in ArcGIS Checkboard pattern

‎11-04-2020

Hello I have replicated the code on this training example https://developers.arcgis.com/python/sample-notebooks/land-cover-classification-using-sparse-training-data/ And are interested in the F1 and mIOU metrics. The example shows F1 metrics which for me came out similar (but not exactly) to the example. They are all in the around 90 mark, all good. My question is when, I run the mIOU score, the scores are 30 or less. Which would seem really low compared to F1. Why are the mIOU scores so low? Many thanks I am running arcgis 1.8.2 and fastai 1.0.60

‎10-06-2020

Hi Brian Yes a AMD Ryzen 9 3950X 16-Core Processor, 3493 Mhz, 16 Core(s), 32 Logical Processor(s) I believe the install was: conda install -c esri -c fastai -c pytorch arcgis pillow scikit-image fastai=1.0.54 pytorch=1.1.0 and the 'conda list' is fastai 1.0.54 py36_1 esri Regards Tim

‎09-06-2020

Thanks Guneet. Yes you are right, thanks for the help. I had tried this before but was getting stopped on the below error (a bug?). ERROR 002860: Tool parameters are inconsistent with the data you are trying to append to. Failed to execute (ExportTrainingDataForDeepLearning). So to resolve this, the "Class Value Field" must contain exactly the same amount of classes as previously run exports in the output directory. Seems kind of odd to me to have this limitation, but thanks again

Online Status	Offline
Date Last Visited	‎03-11-2021 01:31 PM

My Ideas

Latest Contributions by TimGrenside

Re: Classify Pixels Deep Learning Package - Sentinel 2

Re: Classify Pixels Deep Learning Package - Sentinel 2

Re: Classify pixels using deep learning with resnet50, GPU memory issues?

Re: Classify pixels using deep learning with resnet50, GPU memory issues?

Re: Classify pixels using deep learning with resnet50, GPU memory issues?

Re: Classify pixels using deep learning with resnet50, GPU memory issues?

Classify pixels using deep learning with resnet50, GPU memory issues?

Low Intersection over Union (mIOU) metric on Unet Classifier.

Re: FATAL ERROR mkl_vml_avx2.dll

Re: Re-generate esri_accumulated_stats.json for deep learning training