new config.yaml file

[yvr354]

how can i get a new formate of config.yaml file with all data and model and metrics and trainer parameters .

[samet-akcay]

Hi @yvr354, the CLI has --print_config flag can print out the full config to train a model on a certain dataset. For example, if you want to train a Padim model on MVTec dataset, you could do the following:

anomalib train --data anomalib.data.MVTec --model anomalib.models.Padim --print_config > config.yaml  

This will print the full config to the screen, from which you could save it as config.yaml file. You could then use this config to train like;

anomalib train -c config.yaml`

Let me know if this works.

[nguyenanhtuan1008]

@samet-akcay
Sorry, How to use export model in new config? I would like to export it to Torch , openvino

[samet-akcay]

    CLI Usage:
        1. To export as a torch ``.pt`` file you can run the following command.
            ```python
            anomalib export --model Padim --export_mode TORCH --data MVTec
            ```
        2. To export as an ONNX ``.onnx`` file you can run the following command.
            ```python
            anomalib export --model Padim --export_mode ONNX --data Visa --input_size "[256,256]"
            ```
        3. To export as an OpenVINO ``.xml`` and ``.bin`` file you can run the following command.
            ```python
            anomalib export --model Padim --export_mode OPENVINO --data Visa --input_size "[256,256]"
            ```
        4. You can also overrride OpenVINO model optimizer by adding the ``--mo_args.<key>`` arguments.
            ```python
            anomalib export --model Padim --export_mode OPENVINO --data Visa --input_size "[256,256]" \
                --mo_args.compress_to_fp16 False
            ```

[nguyenanhtuan1008]

@samet-akcay
Thank you so much.

[samet-akcay]

Sure, for more details, you could refer to the following documentation
https://anomalib.readthedocs.io/en/latest/markdown/guides/reference/engine/index.html#anomalib.engine.Engine.export