Add back 3rd plot. Accept that it's broken on Firefox

app.py

@@ -298,6 +298,10 @@ The numbers and letters in brackets like `[3.L]` refer to the parent node's posi
 The trees are a lot deeper than what it shows in the small plot.
 Hugging Face makes the plot ENORMOUS if you expand it, so that isn't much help.
 Pick your poison.
+
+Also, it seems to break on Firefox.
+It works locally on Firefox, but it breaks when I look at it on Hugging Face on Firefox.
+🤷
 """)
 
 
@@ -406,15 +410,12 @@ Later on,
 got me out of a jam.
 This `empet` character is pretty much the only one who answers Python posts on Plotly's forums.
 As far as I can tell, that's because they're the only person in the world who understands Plotly's Python library.
-
-I'm glad it was challenging, though.
-I did go into this hoping for something more interesting than a donut plot.
-Maybe I'll think on the `value` and `gain` fields a bit and come up with a version 2.
 """)
 
-  #st.markdown("""
-#This plot is similar to the plot above, but the slider here coordinates with a table of the data I extracted to plot each tree.
-#""")
+  st.markdown('## Check out the data!')
+  st.markdown("""
+This plot is similar to the plot above, but the slider here coordinates with a table of the data I extracted to plot each tree.
+""")
 
   # This works the way I want
   # but the plot is tiny
@@ -430,20 +431,23 @@ Maybe I'll think on the `value` and `gain` fields a bit and come up with a versi
   # so it won't let me cache that function
   # I could pack the dataframe bytes to smuggle them past that check
   # but whatever
-  #idx = st.slider(
-    #label='Which tree do you want to see?',
-    #min_value=0,
-    #max_value=len(figures)-1,
-    #value=0,
-    #step=1
-  #)
-  #st.markdown(f'### Tree {idx}')
-  #st.plotly_chart(figures[idx])
-  #st.dataframe(trees[idx])
-  #st.markdown("""
-#This section is mostly just to warn you against making the same foolhardy decision to marry the innermost guts of SciKit-Learn to the sparsely documented world of Plotly animations in Python.
-#
-#""")
+  idx = st.slider(
+    label='Which tree do you want to see?',
+    min_value=0,
+    max_value=len(figures)-1,
+    value=0,
+    step=1
+  )
+  st.markdown(f'### Tree {idx}')
+  st.plotly_chart(figures[idx])
+  st.dataframe(trees[idx])
+  st.markdown("""
+This section is mostly just to warn you against making the same foolhardy decision to marry the innermost guts of SciKit-Learn to the sparsely documented world of Plotly animations in Python.
+
+I'm glad it was challenging, though.
+I did go into this hoping for something more interesting than a donut plot.
+Maybe I'll think on the `value` and `gain` fields a bit and come up with a version 2.
+""")
 
   # This is still super slow even if it's only showing the dataframes
   # I'm just going to leave it out entirely
@@ -459,16 +463,19 @@ Maybe I'll think on the `value` and `gain` fields a bit and come up with a versi
 
   # Cutting this out fixed the broken animation on Hugging Face
   # So that issue was perf-based
-  st.markdown('#### (secret third plot)')
-  st.markdown("""
-I orginally had a third viz here where you could move a slider to see the data I used to make each plot.
-That viz recalulated every value in the entire app each time the slider moved.
-I had to remove it to get enough perf for the animation to play correctly.
-If you're feeling brave, you can follow the Quickstart in the README to run this app yourself.
-Then you can uncomment that viz to satisfy your curiosity.
-There's definitely some way to fix it.
-Maybe another milestone for v2.
-""")
+  #
+  # The issue is back. My theory now is that it's just a Firefox problem.
+  # works fine on Chromium
+  #st.markdown('#### (secret third plot)')
+  #st.markdown("""
+#I orginally had a third viz here where you could move a slider to see the data I used to make each plot.
+#That viz recalulated every value in the entire app each time the slider moved.
+#I had to remove it to get enough perf for the animation to play correctly.
+#If you're feeling brave, you can follow the Quickstart in the README to run this app yourself.
+#Then you can uncomment that viz to satisfy your curiosity.
+#There's definitely some way to fix it.
+#Maybe another milestone for v2.
+#""")
 
 
 if __name__=='__main__':