def make_plots(csv_file, index_col=None, title=None, cluster_levels=2):
"""
Function for making series of plots.
Parameters
----------
csv_file : str
Path to the csv_file to plot
index_col : str, optional
Name of the index column in csv_file
title : str, optional
Title of the dataset.
cluster_levels : int (default=1)
Number of levels for hierarchial clustering
Returns
-------
out : dict
Dictionary containing html div for each plot
"""
ds = lds.CSVDataSet(csv_file, name=title, index_column=index_col)
#Make plots and save as div
location_heatmap = lpl.LocationHeatmap(
ds, mode='div').plot(showticklabels=True)
location_lines = lpl.LocationLines(ds,
mode='div').plot(showticklabels=True)
#Z-transform the data
ds.D = ds.D.apply(zscore)
#Make plots on transformed data
histogram_heatmap = lpl.HistogramHeatmap(
ds, mode='div').plot(showticklabels=True)
scree_plot = lpl.ScreePlotter(ds, mode='div').plot()
corr_matrix = lpl.CorrelationMatrix(ds,
mode='div').plot(showticklabels=True)
#HGMM plots
cluster_ds = lcl.HGMMClustering(ds, levels=cluster_levels)
hgmm_dendogram = lpl.HGMMClusterMeansDendrogram(cluster_ds,
mode='div').plot()
hgmm_pair_plot = lpl.HGMMPairsPlot(cluster_ds, mode='div').plot()
hgmm_stacked_mean = lpl.HGMMStackedClusterMeansHeatmap(
cluster_ds, mode='div').plot(showticklabels=True)
hgmm_cluster_mean = lpl.HGMMClusterMeansLevelHeatmap(
cluster_ds, mode='div').plot(showticklabels=True)
hgmm_cluster_means = lpl.HGMMClusterMeansLevelLines(
cluster_ds, mode='div').plot(showticklabels=True)
out = {
"Histogram Heatmap": histogram_heatmap,
"Location Heatmap": location_heatmap,
"Location Lines": location_lines,
"Scree Plot": scree_plot,
"Correlation Matrix": corr_matrix,
"Hierarchical GMM Dendogram": hgmm_dendogram,
"Pair Plot": hgmm_pair_plot,
"Cluster Stacked Means": hgmm_stacked_mean,
"Cluster Mean Heatmap": hgmm_cluster_mean,
"Cluster Mean Lines": hgmm_cluster_means
}
return out
lpl.ScreePlotter(EEG_Embedded, mode="savediv", base_path=out_emb_base).plot()
# In[15]:
lpl.HGMMClusterMeansDendrogram(DM, mode="savediv",
base_path=out_base).plot(level=2)
lpl.HGMMClusterMeansDendrogram(EEG_Embedded,
mode="savediv",
base_path=out_emb_base).plot(level=2)
# In[16]:
lpl.HGMMStackedClusterMeansHeatmap(DM, mode="savediv",
base_path=out_base).plot(level=2)
lpl.HGMMStackedClusterMeansHeatmap(EEG_Embedded,
mode="savediv",
base_path=out_emb_base).plot(level=2)
# In[17]:
lpl.HGMMClusterMeansLevelLines(DM, mode="savediv",
base_path=out_base).plot(level=2)
lpl.HGMMClusterMeansLevelLines(EEG_Embedded,
mode="savediv",
base_path=out_emb_base).plot(level=2)
# In[18]:
lpl.ScreePlotter(DM, mode="savediv", base_path=out_base).plot()
lpl.ScreePlotter(EEG_Embedded, mode="savediv", base_path=out_emb_base).plot()
lpl.EigenvectorHeatmap(HBN_Embedded, mode="savediv", base_path=out_base).plot()
# In[ ]:
lpl.HGMMPairsPlot(HBN_Embedded, mode="savediv",
base_path=out_emb_base).plot(level=1)
# In[ ]:
lpl.HGMMClusterMeansDendrogram(HBN_Embedded,
mode="savediv",
base_path=out_emb_base).plot(level=1)
# In[ ]:
lpl.HGMMStackedClusterMeansHeatmap(HBN_Embedded,
mode="savediv",
base_path=out_emb_base).plot(level=4)
# In[ ]:
lpl.HGMMClusterMeansLevelHeatmap(HBN_Embedded,
mode="savediv",
base_path=out_emb_base).plot(level=4)
# In[ ]:
lpl.HGMMClusterMeansLevelLines(HBN_Embedded,
mode="savediv",
base_path=out_emb_base).plot(level=4)