ProGamerGov

I have created PR with the changes. It's not the prettiest installation guide, but it resolves the Tensorflow version related errors people have been having.

Is there a PR in this repo yet for implementing ControlNet v1.1?

> how many false positive and how much complicated making of arts with these kind of filters? this feedback is very welcomed Content filters are fundamentally art incompatible with free...

@AndreiSva @aleqniko They are still working on the model and seem to want the current test version to not become the dominant version used by the community. Once this first...

@jossevessies The default ImageNet values come from the Lucid Tensorflow library, and there's more about how to calculate them here: https://stackoverflow.com/questions/64015444/how-to-calculate-the-3x3-covariance-matrix-for-rgb-values-across-an-image-datase

@vikhyat I did a quick test, and everything appears to be working for batch inference. Thanks for the quick response and update!

You are referring to a batch size greater than 1 to apply each prompt to, correct?

@vikhyat The batch processing from `2024-03-05` with `2024-03-06` seems to be broken, but flash attention seems to work in 06

Using revision `2024-03-06` results in this: ``` Traceback (most recent call last): File "moondream_batch.py", line 201, in run_model captions = model.answer_question(enc_images, [query_prompt] * len(batch_images), tokenizer) File "/home/user/.cache/huggingface/modules/transformers_modules/vikhyatk/moondream2/c064b4c517bbdbbad6351b1959c90b79858822be/moondream.py", line 95, in...

Thanks! Now, I'm trying the new function and it seems to be resulting a new error message ``` captions = model.batch_answer(enc_images, [query_prompt] * len(batch_images), tokenizer) File "/home/user/.cache/huggingface/modules/transformers_modules/vikhyatk/moondream2/c064b4c517bbdbbad6351b1959c90b79858822be/moondream.py", line 120, in...