POST
|
Thank Dan! I actually run ArcPro on a Windows desktop deployed by IT in Azure as I believe. I had to chance to compare but I will ask them tomorrow in the meeting. We are moving forward, not backward and having a 10s script running in 6m is not easy to tolerant 🙂 I did think about upgrade them to Py toolbox. However, it the issue is somewhere in the hardware, I dont think pyt will be better. I hate to move up to creating Addin in .Net because it is a big job... Kind regards, L
... View more
12-05-2021
05:43 PM
|
0
|
1
|
736
|
POST
|
Hi All, I'm now migrating my script tools from ArcMap to ArcPro (2.7.4). I noticed that all the tools run slower than normal. Some of my tools use the Feature Set as Input to get the temporary polygon digitised by users then doing something with the polygon. Old version of the tool in ArcMap takes 10 seconds to run while new version in ArcPro 2.7.4 takes 5 to 6 minutes (!) to run. Has anyone experience the same issue and found the solution? Thank you! Kind regards, L
... View more
12-05-2021
03:52 PM
|
0
|
3
|
771
|
IDEA
|
Thank Nico for the update! Wonder when will we have the Beta to try? And also the function to fly with our own DEM? the general SRTM DEM is not that good in many locations we operate. Kind regards, LN
... View more
11-01-2021
12:24 PM
|
0
|
0
|
1488
|
IDEA
|
Hi Nico, That is the one we need, chaining all small flight plans together and flight. After inspecting the UGCS, I think that function is very useful for us in forestry. We need to move from traditional plotting to drone plotting. Early 2022 is still 6mth... However, we have already subscribed to SiteScan, we need to wait 🙂 Thanks, Lan
... View more
07-13-2021
07:30 PM
|
0
|
0
|
1688
|
IDEA
|
We need to fly over multiple small plots and I know someone can use UGSC to do this job efficiently with only 13 photos taken on each plot, 1 battery (25 min). Photos on each plot then stitched together to create ortho img for each plot. We have access to SiteScan flight planner and hope that it will have this feature soon. Thanks, Lan
... View more
07-08-2021
08:12 PM
|
6
|
11
|
1819
|
POST
|
Thanks Tim! Your answer is so sharp! it shades some lights now. I have a subset of 50cm res image, dimension 5818x4969 (attached images), I tried to get 908 tree samples on that images but they are quite clustered. IF "it's good practise to label every object in your training data" I may have to reduce the dimension of the subset raster and label more trees. Labelling all trees in the 5818x4969 is time consuming :). I will try this option first! Training data export: 64x64 with stride 32x32: 370 imgs, 128x128 stride 64x64: 310imgs, 256x256 stride 128x128: 129imgs. Imagery quality: the subset looks OK at 1:1000 zoom scale (attached). However, at 50cm resolution, we do not expect much. Imagery enhancement: I know that preprocessing/enhancing imagery is good for classification/DL. What would you recommend in term of software/methodology? Thank you very much!
... View more
06-22-2021
06:17 PM
|
0
|
0
|
1653
|
POST
|
Hi Tim, For SSD I had to use chip 64x64, lr0006, 9 epochs and got a good acc score of 0.26. I kept retrain the model with the same lr and number of epochs. However, the second iteration (9 + 9 ep) actually yields lower score (0.17) and the next iteration (9+9+9) is the worst. I have seen the valid_loss when down all the time, wonder why the output is getting worst. I attached the best output I have. I have not tried YOLOv3 yet, maybe today. But I have tried the FasterRCNN yesterday with 64x64 chip, with recommended lr of 00005. However, I could not predict anything, acc score is 0. Interesting. I tried to export chip size of 256 and then 128,. However, ArcPro generated a small number of chips that not even enough to run the data prep. Could be issue with my sampling strategy or the input image need to be preprocessed to improve the sharpness in between objects (trees). Thanks, Lan
... View more
06-22-2021
02:28 PM
|
0
|
0
|
1656
|
POST
|
Hi Tim, Thanks for reply! I have not try that predict() and I will try it now. I'm just stepping into this world and my questions may sounds stupid 🙂 When I saved a model, I found the ArcGISObjectDetection.py file in the model folder. I have tried to search for it but could not find and docs related to this file, what is it and how to use it. None of the tutorials I crossed mentions this python file. Do you have any idea what is for? Thanks! Lan
... View more
06-22-2021
02:15 PM
|
0
|
1
|
800
|
POST
|
Hi All, I'm trying to train SSD model with arcgis.learn to detect trees from aerial photos. After few adjustments, the model was trained with lr 0.0005 and 8 epoches. However, looking at the "average_precision_score", it is only 0.06. Has anyone crossed this issue and how to improve this parameter? And how much is an acceptable score? I have searched for docs related to "average_precision_score" but not much found. Below is the content of my emd. Thank you and kind regards, Lan { "Framework": "arcgis.learn.models._inferencing", "InferenceFunction": "ArcGISObjectDetector.py", "ModelConfiguration": "_DynamicSSD", "ModelType": "ObjectDetection", "ExtractBands": [ 0, 1, 2 ], "backbone": "resnet34", "Grids": [ 4 ], "Zooms": [ 1.0 ], "Ratios": [ [ 1.0, 1.0 ] ], "SSDVersion": 2, "Classes": [ { "Value": 1, "Name": "Tree", "Color": [ 102, 2, 7 ] } ], "ModelFile": "treecount_chip64_lr0005.pth", "ImageHeight": 64, "ImageWidth": 64, "ImageSpaceUsed": "MAP_SPACE", "LearningRate": "5.0000e-04", "ModelName": "SingleShotDetector", "backend": "pytorch", "ModelParameters": { "backbone": "resnet34", "backend": "pytorch" }, "average_precision_score": { "Tree": 0.06539047501100659 }, "resize_to": null, "IsMultispectral": false }
... View more
06-20-2021
03:53 PM
|
0
|
4
|
1736
|
POST
|
Hi All, I could train a SSD model, got the emd and dlpk using arcgis.learn. However, it was not clear in the docs how to run Object detection using pre-trained model with arcgis.learn. I don't have Image Analyst extension so I'm looking for a way to run detection without it? Thank you very much! Kind regards, Lan
... View more
06-20-2021
03:49 AM
|
0
|
3
|
866
|
POST
|
Hi All, I started playing with the AGOL/Dashboard/Advanced Editing/Arcade and found there are some functions not available for me. Could anyone shade some lights why they are not available? Thanks, Lan
... View more
04-26-2021
05:29 PM
|
1
|
5
|
1954
|
POST
|
Hi Craig, Thanks for your answer! I have tried all possibilities I could think of: changing sync time, manual sync, networks.... and experienced that it updates faster when I kill the app and restart. I'm on Note9 Samsung which is a solid device. The classical project with the old app work faster. Thanks,
... View more
10-01-2020
02:39 PM
|
0
|
1
|
2688
|
POST
|
Hi All, I'm on the latest Workforce for Android. It takes ages for the app to update the new Todo list, even with manual hitting the Sync button. I sometimes have to turn off the app and come back to try the Sync and it works. Has anyone suffered the same issue? Thanks,
... View more
09-30-2020
08:34 PM
|
1
|
21
|
5140
|
POST
|
Hi All, I try to convince myself to use ArcPro everyday now. Just one thing I can't figure out: How to quickly add the output from the History into the map ToC? It was so useful to have it in ArcMap Geoprocessing\Results, just right click and add input or output into the ToC. I did not see any way in ArcPro, except for copying the path then Add data using that path. Thanks!
... View more
07-29-2020
05:13 PM
|
0
|
0
|
441
|
POST
|
Hi All, I'm facing an interesting issue. I have a view in SDE which the number of the record changes/increases all the time. I have published it to our enterprise server. Not all records came through from SDE to the Feature service and I had to published it/overwrote the service few times until the records between the number of records match the SDE view. I have found this thread but it is not that close to my issue https://community.esri.com/thread/171089 Thank you All!
... View more
05-24-2020
09:56 PM
|
0
|
0
|
396
|
Title | Kudos | Posted |
---|---|---|
1 | 09-30-2020 08:34 PM | |
6 | 07-08-2021 08:12 PM | |
1 | 04-26-2021 05:29 PM | |
1 | 06-10-2019 04:33 PM | |
2 | 06-11-2019 02:46 PM |
Online Status |
Offline
|
Date Last Visited |
03-30-2022
03:23 PM
|