In AI center, pipeline I am getting failed

I have created 2 folder as Training and evoluation and uploaded into Dataset, and for the ML package I have selected Image Classifier

In the pipeline I am getting below error and its failed

Train only of ImageClassification 4.0 launched - Run 9ed31d36-1724-438e-b063-15e8855cd0f7
Train only of ImageClassification 4.0 started - Run 9ed31d36-1724-438e-b063-15e8855cd0f7
Train only of ImageClassification 4.0 scheduled - Run 9ed31d36-1724-438e-b063-15e8855cd0f7
Train only of ImageClassification 4.0 failed - Run 9ed31d36-1724-438e-b063-15e8855cd0f7

Error Details : Pipeline failed due to ML Package Issue

2024-02-20 08:41:54,391 - UiPath_core.trainer_run:main:83 - INFO: Starting training job…
2024-02-20 08:41:55,326 - matplotlib:_get_config_or_cache_dir:484 - WARNING: Matplotlib created a temporary config/cache directory at /tmp/matplotlib-qlhcly7a because the default path (/home/aicenter/.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.
2024-02-20 08:41:55,782 - matplotlib.font_manager:_load_fontmanager:1443 - INFO: generated new fontManager
2024-02-20 08:41:57,975 - UiPath_core.storage.azure_storage_client:download:118 - INFO: Dataset from bucket folder training-a05cf1db-e98e-46b2-a5b3-4d60f9675158/a79238a7-b9b4-4a7f-a3da-0c2347ffad23/906d7bca-ed84-4b9c-a8dc-1a9a2fc85efa with size 18 downloaded successfully
2024-02-20 08:41:57,975 - UiPath_core.training_plugin:train_model:130 - INFO: Start model training…
2024-02-20 08:41:57,975 - UiPath_core.training_plugin:initialize_model:124 - INFO: Start model initialization…
2024-02-20 08:41:57,978 - UiPath_core.training_plugin:initialize_model:127 - INFO: Model initialized successfully
2024-02-20 08:41:57,978 - UiPath_core.training_plugin:model_run:179 - ERROR: Training failed for pipeline type: TRAIN_ONLY, error: [Errno 2] No such file or directory: ‘/microservice/dataset/images’
2024-02-20 08:41:57,980 - UiPath_core.trainer_run:main:100 - ERROR: Training Job failed, error: [Errno 2] No such file or directory: ‘/microservice/dataset/images’
Traceback (most recent call last):
File “/home/aicenter/.local/lib/python3.8/site-packages/UiPath_core/trainer_run.py”, line 95, in main
wrapper.run()
File “/microservice/training_wrapper.py”, line 58, in run
return self.training_plugin.model_run()
File “/home/aicenter/.local/lib/python3.8/site-packages/UiPath_core/training_plugin.py”, line 195, in model_run
raise ex
File “/home/aicenter/.local/lib/python3.8/site-packages/UiPath_core/training_plugin.py”, line 171, in model_run
self.run_train_only()
File “/home/aicenter/.local/lib/python3.8/site-packages/UiPath_core/training_plugin.py”, line 255, in run_train_only
self.train_model(self.local_dataset_directory)
File “/home/aicenter/.local/lib/python3.8/site-packages/UiPath_core/training_plugin.py”, line 132, in train_model
response = self.model.train(directory)
File “/microservice/train.py”, line 37, in train
self.process_data()
File “/microservice/train.py”, line 61, in process_data
self.df_train, self.df_test = preprocess.preprocess_data(self.opt)
File “”, line 11, in preprocess_data
File “”, line 31, in create_df_from_single_folder
FileNotFoundError: [Errno 2] No such file or directory: ‘/microservice/dataset/images’
2024-02-20 08:41:57,980 - UiPath_core.trainer_run:main:107 - INFO: Job run stopped.

When you go to create a new pipeline run, what values do you input into the configuration screen?

Also, please include screenshots of your ML Packages and ML Skills pages.

For ML Package,

As Pipeline got failed for both, I have not added ML Skills as nothing available as status.
And Pipeline I have selected Train and Evaluation in the Pipeline type and dataset is uploaded folder (Training folder and Evaluation folder, inside subfolder as Images folder and 3 subfolder as Car, bike and bus folder)

You need to deploy the ML Package before you can train it. Go through the process of deploying your ML Package by selecting the ImageClassification package type in your screenshot above and following the steps. Then you will need to wait for it to deploy. Then run your training pipeline on the deployed package.

Same thing I have done.

  1. Upload Dataset(Train and Evaluation folder)
    2.Selected ML Package(ImageClassification)
    3.Came to Pipeline(selected Train and Evaluation Type)

Hi @Sharanabasava ,

Have you tried full pipeline run?

And what is the environment variable passed ?

I have tried Full pipeline, still it got failed
Environment variable changed default 20 to 5, and nothing I have added:
image

Image classification problems training pipeline - Help / AI Center - UiPath Community Forum

Check this .

I have tried same thing,

but it did not work

DO you have license to train the ML (How many AI bots are free)?

I have one deployed on “Text classification”, other one I was trying Image classification

Can you try recreating the project again with same steps ? Or have you tried it.

Tried multiple times, but no success

Try changing the Dataset location used for training.

First I have kept in the folder and tried, not worked.
Then kept Dataset folder in the desktop and tried, but still no

The input dataset has a structure it is required to follow. It needs to be like this:

-<Main Folder That You Select>
     -images
           -Car
                -car1.jpg
                -car2.jpg
                -car3.jpg 
           -Bike
                -bike1.jpg
                -bike2.jpg
                -bike3.jpg 
           -Bus
                -Bus1.jpg
                -Bus2.jpg
                -Bus3.jpg

Make sure this is the folder structure and you are selecting the correct directory and not a parent or child directory. Also make sure there aren’t any random folders or files in your dataset or it will fail.

Please send a screenshot of the failed pipeline settings by clicking into it on the pipelines page. It should look like this:

1 Like

I actually think I see your problem… Your input dataset has a capital ‘I’ in the Images folder name, but it may need to be a lowercase ‘i’ for the model to find it. Try renaming the Images folder to ‘images’ and see if that works.

I have tried this, got failed

Thank you Nathan, Suraj for continuous help. It got resolved
While selecting ML package I have selected 4 instead of 23.2.0.1 version.
After seeing Nathan image, I have tried with 23.2.0.1 this package now both status is successfully

1 Like