perplexity option

Author: mluerig

bioencoder/scripts/interactive_plots.py

@@ -67,6 +67,7 @@ def interactive_plots(
     color_map = hyperparams.get("color_map", "jet")
     plot_style = hyperparams.get("plot_style", 1)
     point_size = hyperparams.get("point_size", 10)
+    perplexity = hyperparams.get("perplexity", None)
 
     ## set up dirs
     data_dir = os.path.join(root_dir,"data",  run_name)
@@ -114,8 +115,10 @@ def interactive_plots(
         ]
         return pd.concat([df, pd.DataFrame(embeddings_train)], axis=1)
 
+    ## reduce dimensionality
+    perplexity = perplexity if perplexity else min(100, len(embeddings_train) // 2)
     reduced_data, colnames, _ = helpers.embbedings_dimension_reductions(
-        embeddings_train
+        embeddings_train, perplexity
     )       
     df = pd.DataFrame(reduced_data, columns=colnames)
     df["paths"] = [ os.path.join("..", "..", item) for item in rel_paths_train]

bioencoder/vis/helpers.py

@@ -214,7 +214,7 @@ def feature_map_normalization(f):
     act_map /= act_map.max()
     return act_map
 
-def embbedings_dimension_reductions(data_table):
+def embbedings_dimension_reductions(data_table, perplexity):
     """
     Perform dimension reduction on the input data.
 
@@ -231,7 +231,7 @@ def embbedings_dimension_reductions(data_table):
     norm_data = (data_table - mean) / std
     pca_obj = decomposition.PCA(n_components=2)
     pca = pca_obj.fit_transform(norm_data)
-    tsne = manifold.TSNE(learning_rate='auto', init='pca').fit_transform(norm_data)
+    tsne = manifold.TSNE(perplexity=perplexity, learning_rate='auto', init='pca').fit_transform(norm_data)
     names = ['PC1', 'PC2', 'tSNE-0', 'tSNE-1']
     return np.hstack((pca, tsne)), names, pca_obj