joenathan commited on
Commit
0052ef8
·
verified ·
1 Parent(s): 27122b9

Give every tab a headline; restore tab 3's section rule; drop the conditional

Browse files

Tabs 2 and 3 each opened with a claim; tab 1 opened with instructions, so the
proposition the other two tabs attack was never stated. Tab 1 now opens with "A
working model that explains itself", making the sequence assert, undermine, what
survives.

Restores the horizontal rule before the conformal section -- present until v3.2,
where it sat where the chart needed to go and was not put back. After a figure, a
table and a blockquote, an h3 alone does not read as a section boundary.

Tab 3 label drops the conditional: "What We'd Show an Auditor" -> "What We Show an
Auditor".

Mirrors llm-wiki commit 18dcbd6.

Files changed (1) hide show
  1. app.py +4 -1
app.py CHANGED
@@ -172,6 +172,7 @@ with gr.Blocks(title="Explainable & Auditable Alloy Defect Detection",
172
  )
173
  with gr.Tab("1 - Classify & Explain"):
174
  gr.Markdown(
 
175
  "Upload a micrograph (or pick an example below). The model predicts the "
176
  "most likely defect class; the heatmap shows which pixels drove that "
177
  "prediction (**Grad-CAM**, implemented from first principles with PyTorch "
@@ -206,7 +207,7 @@ with gr.Blocks(title="Explainable & Auditable Alloy Defect Detection",
206
  )
207
  gr.Markdown(SPURIOUS_MARKDOWN)
208
 
209
- with gr.Tab("3 - What We'd Show an Auditor"):
210
  gr.Markdown(
211
  """
212
  ### First: can we even tell whether the heatmap is faithful?
@@ -240,6 +241,8 @@ with gr.Blocks(title="Explainable & Auditable Alloy Defect Detection",
240
  > why we treat this curve as supporting evidence rather than proof, and why the
241
  > guarantee below is what we would actually put in front of an auditor.
242
 
 
 
243
  ### So what survives? Conformal Risk Control
244
 
245
  **Not covered anywhere in the course** — taken from Shen & Liu,
 
172
  )
173
  with gr.Tab("1 - Classify & Explain"):
174
  gr.Markdown(
175
+ "### A working model that explains itself\n\n"
176
  "Upload a micrograph (or pick an example below). The model predicts the "
177
  "most likely defect class; the heatmap shows which pixels drove that "
178
  "prediction (**Grad-CAM**, implemented from first principles with PyTorch "
 
207
  )
208
  gr.Markdown(SPURIOUS_MARKDOWN)
209
 
210
+ with gr.Tab("3 - What We Show an Auditor"):
211
  gr.Markdown(
212
  """
213
  ### First: can we even tell whether the heatmap is faithful?
 
241
  > why we treat this curve as supporting evidence rather than proof, and why the
242
  > guarantee below is what we would actually put in front of an auditor.
243
 
244
+ ---
245
+
246
  ### So what survives? Conformal Risk Control
247
 
248
  **Not covered anywhere in the course** — taken from Shen & Liu,