linfa
linfa copied to clipboard
rust-ml
A Rust machine learning framework.
Hello! Would you be interested in MeanShift implementation? I'm planning on using it as clustering algorithm in my project, and would like to contribute to Linfa as well 🦀
Hi, I want to get loss for each iteration (not the whole loss of the model) when using logistic regression, this is what I really want:  how to get...
Hi, who can kindly tell the meaning of "gradient_tolerance" and "alpha" parameters ? what's the default value for them ? which one is the learning rate ? I cannot not...
Hi, here is a confusion matrix getting from logistic regression: |classes | negative | positive| |---|---|---| |negative | 295 | 0 | |positive | 20 | 4 | But I...
Hi, We could get the prediction of logisticregression model using ``` let validation = model.set_threshold(threshold).predict(&dataset_validation); ``` However, can we get the prediction score of each records ? Then we can...
Currently i'm building a wasm project that will expose some clustering functionality to the browser. Questions: - Do we have grid search functionality? Or should I simply loop over multiple...
SVM(Multi) calculation consumes memory and time, so I want to save the model calculation data for subsequent calls, but it seems that SVM does not support the serde feature. How...
Linfa supports pca and tsne now. Uniform Manifold Approximation and Projection (UMAP) is a dimension reduction technique that can be used for visualisation similarly to t-SNE, but also for general...
The [DBSCAN in python sklearn](https://scikit-learn.org/stable/modules/generated/sklearn.cluster.DBSCAN.html#sklearn.cluster.DBSCAN) returns the indices of core samples. It can be beneficial to find representative samples. Will the DBSCAN in linfa have a similar feature?
Firstly, let me say I'm _very_ new to data science / ML so my understanding / terminology may be wrong. Please bare with me, thanks in advance. I'm using a...
