Improving the t-sne plot

src/nhssynth/modules/plotting/plots.py

@@ -27,25 +27,25 @@ def factorize_all_categoricals(
 
 
 def tsne(
-    X_gt: pd.DataFrame,
-    X_syn: pd.DataFrame,
+    real: pd.DataFrame,
+    synth: pd.DataFrame,
 ) -> None:
-    tsne_gt = TSNE(n_components=2, random_state=0, learning_rate="auto", init="pca")
-    proj_gt = pd.DataFrame(tsne_gt.fit_transform(factorize_all_categoricals(X_gt)))
+    tsne_real = TSNE(n_components=2, init="pca")
+    proj_real = pd.DataFrame(tsne_real.fit_transform(factorize_all_categoricals(real)))
 
-    tsne_syn = TSNE(n_components=2, random_state=0, learning_rate="auto", init="pca")
-    proj_syn = pd.DataFrame(tsne_syn.fit_transform(factorize_all_categoricals(X_syn)))
+    tsne_synth = TSNE(n_components=2, init="pca")
+    proj_synth = pd.DataFrame(tsne_synth.fit_transform(factorize_all_categoricals(synth)))
 
     fig = go.Figure()
 
-    fig.add_scatter(x=proj_gt[0], y=proj_gt[1], mode="markers", marker=dict(size=5), opacity=0.75, name="Real data")
+    fig.add_scatter(x=proj_real[0], y=proj_real[1], mode="markers", marker=dict(size=5), opacity=0.75, name="Real data")
     fig.add_scatter(
-        x=proj_syn[0], y=proj_syn[1], mode="markers", marker=dict(size=5), opacity=0.75, name="Synthetic data"
+        x=proj_synth[0], y=proj_synth[1], mode="markers", marker=dict(size=5), opacity=0.75, name="Synthetic data"
     )
 
     # Set axis labels and legend
     fig.update_layout(
-        title="t-SNE plot",
+        title="t-SNE Plot",
         xaxis_title="t-SNE 1",
         yaxis_title="t-SNE 2",
         legend=dict(x=0, y=1, bgcolor="rgba(255, 255, 255, 0.5)"),