search

sidhomj / DeepTCR

Deep Learning Methods for Parsing T-Cell Receptor Sequencing (TCRSeq) Data
https://sidhomj.github.io/DeepTCR/
MIT License
113 stars 40 forks source link

VAE AUC violin plot Y-axis value >1 #3

Closed hejing3283 closed 5 years ago

hejing3283 commented 5 years ago

I am running the unsupervised tutorial by

Instantiate training object

DTCRU = DeepTCR_U('Tutorial')

Load Data from directories

DTCRU.Get_Data(directory='github/DeepTCR/Data/Rudqvist',Load_Prev_Data=False,aggregate_by_aa=True, aa_column_beta=1,count_column=2,v_beta_column=7,d_beta_column=14,j_beta_column=21)

Train VAE

DTCRU.Train_VAE(Load_Prev_Data=False,accuracy_min=0.9)

Output-----------

Finding 30 nearest neighbors using minkowski metric and 'auto' algorithm Neighbors computed in 50.87618613243103 seconds Jaccard graph constructed in 12.015304803848267 seconds Wrote graph to binary file in 3.909883975982666 seconds Running Louvain modularity optimization After 1 runs, maximum modularity is Q = 0.991586 Louvain completed 21 runs in 9.033319234848022 seconds PhenoGraph complete in 75.99179887771606 seconds Clustering Done AUC__1

sidhomj commented 5 years ago

violin plots model the distribution of the underlying data. The actual points are not greater than 1. You can change the plot type to a different one like boxplot or swarm plot.

hejing3283 commented 5 years ago

In this case I think violin plot is not a good choice for visualization. Thanks a lot!

sidhomj commented 5 years ago

In that case, change the plot_type parameter:

    plot_type: str
        Type of plot as taken by seaborn.catplot for kind parameter:
        options include (strip,swarm,box,violin,boxen,point,bar,count)