Выполнение завершено без обученной модели

[irchimi]

Ошибка: в System.Linq.Enumerable.Select[TSource,TResult](IEnumerable1 source, Func2 selector) в Microsoft.ML.ModelBuilder.AutoMLService.Experiments.AutoMLExperiment`3.d23.MoveNext() в E:\A_work\1326\s\Microsoft.ML.ModelBuilder.AutoMLService\Experiments\AutoMLExperiment.cs:строка 117 --- Конец трассировка стека из предыдущего расположения, где возникло исключение --- в System.Runtime.ExceptionServices.ExceptionDispatchInfo.Throw() в System.Runtime.CompilerServices.TaskAwaiter.HandleNonSuccessAndDebuggerNotification(Task task) в Microsoft.ML.ModelBuilder.AutoMLEngine.d28.MoveNext() в E:\A_work\1326\s\Microsoft.ML.ModelBuilder.AutoMLService\AutoMLEngineService\AutoMLEngine.cs:строка 134

Вариант: Классификация изображений Набор данных: Три папки для классификации по 12 картинок в каждой

[LittleLittleCloud]

Можете ли вы попытаться увеличить время тренировки?

[irchimi]

В варианте Image Classification нет выбора времени тренировки. Соответственно, после запуска тренировки через вариант Local - процесс идет какое-то время и выдает ошибку, которую я описала выше. Скрины приложу ниже.

[LittleLittleCloud]

Did you see the log start with [Source=ImageClassificationTrainer; before training canceled? (sorry my Russian is bad,,,, so switch back to English here)

|     Trainer                              MicroAccuracy  MacroAccuracy  Duration #Iteration                     |
[Source=ImageClassificationTrainer; ImageClassificationTrainer, Kind=Trace] Channel started
[Source=ImageClassificationTrainer; ImageClassificationTrainer, Kind=Trace] Channel started
[Source=ImageClassificationTrainer; Ensuring meta files are present., Kind=Trace] Channel started
[Source=ImageClassificationTrainer; Ensuring meta files are present., Kind=Info] Downloading resnet_v2_50_299.meta from https://aka.ms/mlnet-resources/resnet_v2_50_299.meta to C:\Users\xiaoyuz\AppData\Local\Temp\MLNET\resnet_v2_50_299.meta
[Source=ImageClassificationTrainer; Ensuring meta files are present., Kind=Info] resnet_v2_50_299.meta: Downloaded 3711 bytes out of 102616931
[Source=ImageClassificationTrainer; Ensuring meta files are present., Kind=Info] resnet_v2_50_299.meta: Downloaded 10264576 bytes out of 102616931
[Source=ImageClassificationTrainer; Ensuring meta files are present., Kind=Info] resnet_v2_50_299.meta: Downloaded 20525056 bytes out of 102616931
[Source=ImageClassificationTrainer; Ensuring meta files are present., Kind=Info] resnet_v2_50_299.meta: Downloaded 30785536 bytes out of 102616931
[Source=ImageClassificationTrainer; Ensuring meta files are present., Kind=Info] resnet_v2_50_299.meta: Downloaded 41050112 bytes out of 102616931
[Source=ImageClassificationTrainer; Ensuring meta files are present., Kind=Info] resnet_v2_50_299.meta: Downloaded 51310592 bytes out of 102616931
[Source=ImageClassificationTrainer; Ensuring meta files are present., Kind=Info] resnet_v2_50_299.meta: Downloaded 61571072 bytes out of 102616931
[Source=ImageClassificationTrainer; Ensuring meta files are present., Kind=Info] resnet_v2_50_299.meta: Downloaded 71835648 bytes out of 102616931
[Source=ImageClassificationTrainer; Ensuring meta files are present., Kind=Info] resnet_v2_50_299.meta: Downloaded 82096128 bytes out of 102616931
[Source=ImageClassificationTrainer; Ensuring meta files are present., Kind=Info] resnet_v2_50_299.meta: Downloaded 92356608 bytes out of 102616931
[Source=ImageClassificationTrainer; Ensuring meta files are present., Kind=Info] resnet_v2_50_299.meta: Downloaded 102616931 bytes out of 102616931
[Source=ImageClassificationTrainer; Ensuring meta files are present., Kind=Info] resnet_v2_50_299.meta: Download complete
[Source=ImageClassificationTrainer; Ensuring meta files are present., Kind=Trace] Channel finished. Elapsed 00:01:12.1951597.
[Source=ImageClassificationTrainer; Ensuring meta files are present., Kind=Trace] Channel disposed
[Source=ImageClassificationTrainer; ImageClassificationTrainer, Kind=Trace] Phase: Bottleneck Computation, Dataset used:      Train, Image Index:   1
[Source=ImageClassificationTrainer; ImageClassificationTrainer, Kind=Trace] Phase: Bottleneck Computation, Dataset used:      Train, Image Index:   2
[Source=ImageClassificationTrainer; ImageClassificationTrainer, Kind=Trace] Phase: Bottleneck Computation, Dataset used:      Train, Image Index:   3
[Source=ImageClassificationTrainer; ImageClassificationTrainer, Kind=Trace] Phase: Bottleneck Computation, Dataset used:      Train, Image Index:   4
[Source=ImageClassificationTrainer; ImageClassificationTrainer, Kind=Trace] Phase: Bottleneck Computation, Dataset used:      Train, Image Index:   5
[Source=ImageClassificationTrainer; ImageClassificationTrainer, Kind=Trace] Phase: Bottleneck Computation, Dataset used:      Train, Image Index:   6
[Source=ImageClassificationTrainer; ImageClassificationTrainer, Kind=Trace] Phase: Bottleneck Computation, Dataset used:      Train, Image Index:   7
[Source=ImageClassificationTrainer; ImageClassificationTrainer, Kind=Trace] Phase: Bottleneck Computation, Dataset used:      Train, Image Index:   8
[Source=ImageClassificationTrainer; ImageClassificationTrainer, Kind=Trace] Phase: Bottleneck Computation, Dataset used:      Train, Image Index:   9
[Source=ImageClassificationTrainer; ImageClassificationTrainer, Kind=Trace] Phase: Bottleneck Computation, Dataset used:      Train, Image Index:  10
[Source=ImageClassificationTrainer; ImageClassificationTrainer, Kind=Trace] Phase: Bottleneck Computation, Dataset used:      Train, Image Index:  11
[Source=ImageClassificationTrainer; ImageClassificationTrainer, Kind=Trace] Phase: Bottleneck Computation, Dataset used:      Train, Image Index:  12
[Source=ImageClassificationTrainer; ImageClassificationTrainer, Kind=Trace] Phase: Bottleneck Computation, Dataset used:      Train, Image Index:  13
[Source=ImageClassificationTrainer; ImageClassificationTrainer, Kind=Trace] Phase: Bottleneck Computation, Dataset used:      Train, Image Index:  14
[Source=ImageClassificationTrainer; ImageClassificationTrainer, Kind=Trace] Phase: Bottleneck Computation, Dataset used:      Train, Image Index:  15
[Source=ImageClassificationTrainer; ImageClassificationTrainer, Kind=Trace] Phase: Bottleneck Computation, Dataset used:      Train, Image Index:  16
[Source=ImageClassificationTrainer; ImageClassificationTrainer, Kind=Trace] Phase: Bottleneck Computation, Dataset used:      Train, Image Index:  17

[irchimi]

My English isn't that good, but I'll try. So far, I have solved the problem by taking the folder out of the project. After that, the learning process ceased to interrupt. "Did you see the log start with [Source=ImageClassificationTrainer; before training canceled?" - No, there was no logbook. The program started training and finished the process with the result "0%" and displayed an error.

Why it depends on changing the position of the folder - I do not understand yet. What does "increase the training time" mean? I know that in some methods you can set the training time. In this method - I did not find it.

[LittleLittleCloud]

What does "increase the training time" mean? I know that in some methods you can set the training time. In this method - I did not find it.

This is because image classification's trainer will determine it's training time by itself, so you don't have to set it. The last time I said to increase training time is because my Russia is terrible... so I didn't know that you are using image classificaiton.

Why it depends on changing the position of the folder - I do not understand yet.

Me neither,,, I will try if I can reproduce it

[LittleLittleCloud]

I'm unable to reproduce it on my PC. my folder structure is like this

where image folder is fruits, which has this folder structure

Can you provide more detailed info,,,, like the folder structure for your project? Thanks

[LittleLittleCloud]

@KotoRina I'm going to close this issue, feel free to reopen it if you still have questions