DFKI-NLP/thermostat issues and pull requests

#18 - Edit dependency to new scikit-learn nomenclature

Pull Request - State: closed - Opened by g8a9 over 1 year ago - 2 comments

#17 - added missing return statement

Pull Request - State: closed - Opened by MNasert over 2 years ago - 1 comment

#16 - Final pull request for release 1.0.2

Pull Request - State: closed - Opened by aj280192 over 2 years ago

#15 - Adding layer deeplift shap and layer gradient shap explainers.

Pull Request - State: closed - Opened by aj280192 over 2 years ago - 1 comment

#14 - adding LIME explainer built with 100 samples

Pull Request - State: closed - Opened by aj280192 over 2 years ago - 1 comment

#13 - Prediction Values in Dataset

Issue - State: closed - Opened by tpatzelt about 3 years ago - 3 comments

#12 - Improve heatmap visualization options (more color schemes, better readability)

Issue - State: open - Opened by nfelnlp over 3 years ago

#11 - Missing assertion that num_labels in dataset corresponds to classification head shape of model

Issue - State: open - Opened by nfelnlp over 3 years ago

#10 - MNLI and XNLI downstream model performance very low

Issue - State: closed - Opened by nfelnlp over 3 years ago - 1 comment
Labels: invalid

#9 - LIME token similarity kernel for input.shape[0] > 1

Issue - State: closed - Opened by nfelnlp over 3 years ago - 3 comments
Labels: help wanted

#8 - AttributeError: 'XLNetModel' object has no attribute 'embeddings'

Issue - State: closed - Opened by nfelnlp over 3 years ago - 6 comments

#7 - batch dimension ignored

Issue - State: closed - Opened by rbtsbg over 3 years ago - 4 comments

#6 - Find out if the cast to long for batch components in forward functions is necessary

Issue - State: open - Opened by nfelnlp over 3 years ago - 1 comment
Labels: question

#5 - [InputXGradient] RuntimeError: One of the differentiated Tensors does not require grad

Issue - State: closed - Opened by nfelnlp over 3 years ago - 2 comments
Labels: bug

#4 - [KernelShap] ZeroDivisionError: Weights sum to zero, can't be normalized

Issue - State: open - Opened by nfelnlp over 3 years ago - 1 comment
Labels: bug

#3 - LIME : "normal_kernel_cuda" not implemented for 'Long'

Issue - State: closed - Opened by nfelnlp over 3 years ago - 4 comments
Labels: bug

#2 - ExplainerCaptum.get_inputs_and_additional_args and .get_forward_function need to be extensible

Issue - State: closed - Opened by nfelnlp almost 4 years ago - 1 comment
Labels: enhancement

#1 - Selection of experiment_in unsafe

Issue - State: closed - Opened by nfelnlp almost 4 years ago - 1 comment
Labels: invalid

Ecosyste.ms: Issues

GitHub / DFKI-NLP/thermostat issues and pull requests