Artem

Hi @annahedstroem, I'd like to work on this feature

So, as per official documentation: A SavedModel contains a complete TensorFlow program, including trained parameters (i.e, [tf.Variable] (https://www.tensorflow.org/api_docs/python/tf/Variable)s) and computation. It does not require the original model building code to...

> OK, we need to discuss the performance gains / pros and cons with jax, separately as discussed on 08.09.22, no `jax` for now. I've changed the implementation to use...

So, I installed `pytest-dead-fixture` from source. Not this is what I get as an output: ```INTERNALERROR> Traceback (most recent call last): INTERNALERROR> File "/Users/artemsereda/anaconda3/envs/Quantus-3.9/lib/python3.9/site-packages/_pytest/main.py", line 269, in wrap_session INTERNALERROR> session.exitstatus...

It is easier to open a new one

Hi @sltzgs, could you please post a full stacktrace of the exception? It would also help if you could provide a code snippet, for which it occurred, and describe your...

> That said, its pretty rare to use `nn.jit`. Usually you just use `jax.jit` over the `train_step` as shown in the [Quick Start](https://flax.readthedocs.io/en/latest/getting_started.html). Hi @cgarciae, it looks like I had...

Hi, so, here [](https://colab.research.google.com/drive/1yQ-RwL1rHmrOtnRbuNYT_pCw9jdV7DXV?usp=sharing) is a notebook demonstrating how NLP explanation evaluation with Quantus could look like in (near) future.

In the first iteration, the goal is to support models from Hugging face Hub with scope limited to text-classification tasks.

Alternative suggestion, we could make PRNG seed library-wide configuration and set it on module load.