it is creating 3 folders but data is contained only by two document and metadata
prediction folder is blank.
how i can my ML model to go through continues training.
Hi @H_khot ,
Could you check if you have the Alias name provided when configuring the Extractors ?
If not, could you check by adding them ? The same needs to be done using the Train Extractors Scope.
Yes you were right, Now it is providing predictions also
I am passed the same Framework Alias to both ML extractor and Trainer
@supermanPunch Can you please help me in How i can make a continues learning ML model.
I have exported that folder that contains 3 folder (document, metadata, predictions).
created train pipeline for training the ML skill based on new files.
but it is throwing error. ML package is not proper.
@H_khot ,
Could you provide us a Screenshot of the Configuration done for the Training Pipeline ?
Also, could you provide more details on the ML Logs generated ?
@supermanPunch Thanku for Helping in Advance.
in my machine it creates a folder with name MachineLearningExtractorTrainer
it directly uploads that document to Cloud Ui path
with the same folder in it
the error it is throwing:
Train only of InvoiceExtractionMLPackage 23.4.1.1 launched - Run 87be00d6-dca9-4bd7-9d6b-f4149d99eadc
Train only of InvoiceExtractionMLPackage 23.4.1.1 started - Run 87be00d6-dca9-4bd7-9d6b-f4149d99eadc
Train only of InvoiceExtractionMLPackage 23.4.1.1 scheduled - Run 87be00d6-dca9-4bd7-9d6b-f4149d99eadc
Train only of InvoiceExtractionMLPackage 23.4.1.1 failed - Run 87be00d6-dca9-4bd7-9d6b-f4149d99eadc
Error Details : Pipeline failed due to ML Package Issue
2023-07-29 06:06:29,995 - UiPath_core.trainer_run:main:74 - INFO: Starting training job…
2023-07-29 06:06:32,455 - matplotlib:_get_config_or_cache_dir:526 - WARNING: Matplotlib created a temporary config/cache directory at /tmp/matplotlib-f96glnr9 because the default path (/.config/matplotlib) is not a writable directory; it is highly recommended to set the MPLCONFIGDIR environment variable to a writable directory, in particular to speed up the import of Matplotlib and to better support multiprocessing.
2023-07-29 06:06:32,645 - matplotlib.font_manager:_load_fontmanager:1544 - INFO: generated new fontManager
2023-07-29 06:06:33,897 - UiPath_core.storage.azure_storage_client:download:118 - INFO: Dataset from bucket folder training-cfef20ef-6d20-457d-91aa-70327e3e1236/69aa9f81-a0cb-433a-9c3b-4786c091e196/c56970c1-e14f-458e-91f5-d39ae7ba2748/MachineLearningExtractorTrainer with size 4 downloaded successfully
2023-07-29 06:06:33,897 - UiPath_core.training_plugin:train_model:129 - INFO: Start model training…
2023-07-29 06:06:33,897 - UiPath_core.training_plugin:initialize_model:123 - INFO: Start model initialization…
2023-07-29 06:06:33,898 - root:initialize_package:195 - INFO: Using package type provided by runtime argument with value: invoices
2023-07-29 06:06:33,898 - root:initialize_package:204 - INFO: Initializing invoices package options …
2023-07-29 06:06:33,899 - root:_valid_doctype_folder_structure:98 - ERROR: images/ does not exist / is empty for invoices dataset
2023-07-29 06:06:33,899 - UiPath_core.training_plugin:model_run:189 - ERROR: Training failed for pipeline type: TRAIN_ONLY, error: Document type invoices not valid, check that document type data is in dataset folder and follows folder structure.
2023-07-29 06:06:33,900 - UiPath_core.trainer_run:main:91 - ERROR: Training Job failed, error: Document type invoices not valid, check that document type data is in dataset folder and follows folder structure.
Traceback (most recent call last):
File “/model/bin/UiPath_core/trainer_run.py”, line 86, in main
wrapper.run()
File “/workspace/model/microservice/training_wrapper.py”, line 64, in run
return self.training_plugin.model_run()
File “/model/bin/UiPath_core/training_plugin.py”, line 205, in model_run
raise ex
File “/model/bin/UiPath_core/training_plugin.py”, line 181, in model_run
self.run_train_only()
File “/model/bin/UiPath_core/training_plugin.py”, line 268, in run_train_only
score = self.train_model(self.local_dataset_directory)
File “/model/bin/UiPath_core/training_plugin.py”, line 131, in train_model
response = self.model.train(directory)
File “/model/bin/UiPath_core/training_plugin.py”, line 119, in model
self.initialize_model()
File “/model/bin/UiPath_core/training_plugin.py”, line 125, in initialize_model
self._model = train.Main()
File “/workspace/model/microservice/train.py”, line 21, in init
self.opt = package_util.initialize_package(args)
File “”, line 206, in initialize_package
File “”, line 144, in get_package_opt
File “”, line 78, in configure_pipeline_options
File “”, line 139, in configure_options
Exception: Document type invoices not valid, check that document type data is in dataset folder and follows folder structure.
2023-07-29 06:06:33,901 - UiPath_core.trainer_run:main:98 - INFO: Job run stopped.
@H_khot ,
Since the Data is directly uploaded, we would need to keep in mind two things.
- Exporting the fine tune Folder into the Export Folder
- Selecting the Right Folder in the Dataset
For the First part, would recommend you to check the below post :
In the Data Manager - Scheduled Exports feature, Second Paragraph, it talks about how the export happens with the Data that needs to re-trained.
Next, we would need to Select the export folder itself for the Pipeline Re-training, since after performing the export, all the data with the fine tune data will be present in the export folder.
Hope the above explanation /points is understandable.
The Below video was very helpful For Knowing About Fine-tune folder & Training ML Model.
1 Like
This topic was automatically closed 3 days after the last reply. New replies are no longer allowed.
ashokkarale
Yoichi
singh_sumit
sven.wullum1
adi.mehare
ppr
sonaliaggarwal47
shahidh.aqeel.shahul
Akash_Javalekar1
Sami_Rajput