How to run the analyzer with models in custom folder and not .cache #265
Unanswered
fadilparves
asked this question in
Q&A
Replies: 1 comment
-
There is an environment variable I personally haven't tried if that works seamlessly. I should. But let me know, if there is anything that is behind your expectation. Note however, that you can only specify the base directory. The sub-directories for weights, configs and datasets are fixed for now. As a side note and because I have seen this before: The weights you specify must not be a path, but a registered model, i.e. it must be one of the model in
|
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Hi everyone,
I have moved all the models and config from the .cache folder to other custom folder, how do I make the analyzer work. For example below is how I define the config_overwrite
config_overwrite = [f"PT.LAYOUT.WEIGHTS_TS={base}/dd_cache/weights/layout/d2_model_0829999_layout_inf_only.ts", f"PT.ITEM.WEIGHTS_TS={base}/dd_cache/weights/item/d2_model_1639999_item_inf_only.ts", f"PT.CELL.WEIGHTS_TS={base}/dd_cache/weights/cell/d2_model_1849999_cell_inf_only.ts", f"OCR.CONFIG.TESSERACT={base}/dd_cache/configs/dd/conf_tesseract.yaml"]
And I am just using the default model that were downloaded by dd during the first run it just that I put them in different folder. But when I run analyzer = dd.get_dd_analyzer(config_overwrite=config_overwrite)
I get the error below
Appreciate if anyone knows how to make this work. Thank you
Beta Was this translation helpful? Give feedback.
All reactions