Select to view content in your preferred language

Error with Train Deep learning Model

550
2
09-26-2023 01:50 PM
NIKOLAYGK
New Contributor

Hello. I am using ArcGIS Pro 3.0 to perform image analysis using the object detection model. At the stage of training the deep learning model, I encountered the following problem, I hope someone will tell me how to solve it.

Traceback (most recent call last):
File "c:\users\Николай\appdata\local\programs\arcgis\pro\Resources\ArcToolbox\toolboxes\Image Analyst Tools.tbx\TrainDeepLearningModel.tool\tool.script.execute.py", line 390, in <module>
execute()
File "c:\users\Николай\appdata\local\programs\arcgis\pro\Resources\ArcToolbox\toolboxes\Image Analyst Tools.tbx\TrainDeepLearningModel.tool\tool.script.execute.py", line 334, in execute
training_model_object.fit(
File "C:\Users\Николай\AppData\Local\Programs\ArcGIS\Pro\bin\Python\envs\arcgispro-py3\Lib\site-packages\arcgis\learn\models\_arcgis_model.py", line 902, in fit
lr = self.lr_find(allow_plot=False)
File "C:\Users\Николай\AppData\Local\Programs\ArcGIS\Pro\bin\Python\envs\arcgispro-py3\Lib\site-packages\arcgis\learn\models\_arcgis_model.py", line 721, in lr_find
raise e
File "C:\Users\Николай\AppData\Local\Programs\ArcGIS\Pro\bin\Python\envs\arcgispro-py3\Lib\site-packages\arcgis\learn\models\_arcgis_model.py", line 718, in lr_find
self.learn.lr_find()
File "C:\Users\Николай\AppData\Local\Programs\ArcGIS\Pro\bin\Python\envs\arcgispro-py3\Lib\site-packages\fastai\train.py", line 40, in lr_find
epochs = int(np.ceil(num_it/len(learn.data.train_dl))) * (num_distrib() or 1)
ZeroDivisionError: division by zero

Failed script (null)...
Failed to execute (TrainDeepLearningModel).

 

Below are screenshots that can point out my mistakes and help solve my problem.

NIKOLAYGK_0-1695760728708.pngNIKOLAYGK_1-1695760829801.png

 

NIKOLAYGK_2-1695761181376.pngNIKOLAYGK_3-1695761196102.pngNIKOLAYGK_4-1695761234194.png

 

0 Kudos
2 Replies
YuriPotawsky
Esri Contributor

Hey @NIKOLAYGK,  Looks like you are utilizing the SSD model.  Do you have any luck running against an RCNN or RetinaNet model? Some of those will be dependent upon the export format for the training data. You might also try lower the max number of epochs as it appears the error is related to that parameter.

Yuri

0 Kudos
NIKOLAYGK
New Contributor

Hi, I tried to use other types of model, as well as reduce the maximum number of epochs, it did not bring me success. In any case, this error remains unchanged. I don't know how to fix it.

0 Kudos