alibi-detect
alibi-detect copied to clipboard
SeldonIO
Algorithms for outlier, adversarial and drift detection
Currently we call kernels as `kernel(x, y)` however models defined using keras's functional API require multiple inputs to be passed as a list. In these cases we instead need to...
The main docs say that only `mse` is supported, but the API docs say that both `mse` and `proba` are supported. Edit: I see in the source code it is...
Some of the models fetched with `fetch_tf_model` are not compatible across python versions different from that the model was trained on. For example, the following code from ``` dataset =...
A number of the example notebooks have very minor bits of deprecated code. Testing the examples during CI would help pick up these issues. However, some of the notebooks involve...
Combination of `preprocess_x_ref=False` and `configure_kernel_from_x_ref=True` is invalid when `preprocess_fn` is not `None`.
This PR changes the format we use to serialize TensorFlow models from the old HDF5 to the newer [SavedModel](https://www.tensorflow.org/guide/saved_model) format. ## Motivation As well as being the default (and recommended)...
Resolves #626. Went straight to 1.1 as [they have nice things](https://github.com/facebook/prophet/releases): - No dependency on `pystan` due to stopped development, the `stan` backend used now is `cmdstanpy` - Wheels provided...
Hi there, I've been exploring data drift detection and have been wanting to test how good alibi is at determining how much a given dataset has drifted. However, my main...
I was executing MMD Drift detector with pytorch backend and while saving the detector i ended up with an error ```python --------------------------------------------------------------------------- NotImplementedError Traceback (most recent call last) /tmp/ipykernel_2460/896443056.py in...
