Tim Kellogg<p>there’s a lot of really cool stuff in <a href="https://hachyderm.io/tags/MLEngineering" class="mention hashtag" rel="tag">#<span>MLEngineering</span></a> that amounts to “train another model”. like using <a href="https://hachyderm.io/tags/SHAP" class="mention hashtag" rel="tag">#<span>SHAP</span></a> to automate feature selection (first you have to train a model though). or <a href="https://hachyderm.io/tags/ConceptSHAP" class="mention hashtag" rel="tag">#<span>ConceptSHAP</span></a> where you train simple linear models on the output of each neural net layer. or anomaly detection, or autoencoders, or...</p>