kevinhug commited on
Commit
553ab50
·
1 Parent(s): d4a8b37
Files changed (1) hide show
  1. app.py +14 -10
app.py CHANGED
@@ -225,13 +225,12 @@ With no need for jargon, SSDS delivers tangible value to our fintech operations.
225
 
226
  gr.Markdown("""
227
 
228
- **Context:**
229
- This analysis is derived from an XGBoost regression model designed to predict house prices. The model utilizes features such as **dist_subway, age, lat, long,** and **dist_stores**.
230
 
231
- Full dataset at the bottom of this tab
232
 
 
 
233
  Explain by Context
234
- ===============
235
  - Sometimes, understanding why an individual defaults requires shifting to a credit-healthy background, altering the baseline E[f(x) | credit healthy] using interventional feature perturbation ([source](https://arxiv.org/pdf/2006.16234.pdf)).
236
 
237
  [UCI Machine Learning Repository - Credit Default Dataset](https://www.kaggle.com/datasets/uciml/default-of-credit-card-clients-dataset)
@@ -249,10 +248,15 @@ f(x) in probability for logistic regression objective using XGBoost
249
  - LIMIT_BAL signifies the amount of given credit in NT dollars (includes individual and family/supplementary credit).
250
  - BILL_AMT1 indicates the bill statement amount in September, 2005 (NT dollar).
251
 
252
- This approach leverages context-specific explanations within credit health parameters to reveal why individuals default, providing valuable insights into repayment behavior and financial health.
253
-
 
 
 
 
 
254
  Explain by Dataset
255
- ===============
256
  - Below are explanation in typical background E[f(x)]
257
 
258
  ![Summary](file=./xgb/data.png)
@@ -265,7 +269,7 @@ Explain by Dataset
265
 
266
 
267
  Explain by Feature
268
- ===============
269
  ![Partial Dependence](file=./xgb/feature.png)
270
 
271
  **Observations:**
@@ -274,7 +278,7 @@ Explain by Feature
274
 
275
 
276
  Explain by Record
277
- ===============
278
  ![Force](file=./xgb/record.png)
279
 
280
  **Contribution to Price:**
@@ -282,7 +286,7 @@ Explain by Record
282
  - **Age** follows as the second significant contributor.
283
 
284
  Explain by Instance
285
- ===============
286
  ![Dependence](file=./xgb/instance.png)
287
 
288
  **Insights:**
 
225
 
226
  gr.Markdown("""
227
 
 
 
228
 
 
229
 
230
+ CREDIT DEFAULT RISK INTERPRETATION
231
+ =======================
232
  Explain by Context
233
+ ----------
234
  - Sometimes, understanding why an individual defaults requires shifting to a credit-healthy background, altering the baseline E[f(x) | credit healthy] using interventional feature perturbation ([source](https://arxiv.org/pdf/2006.16234.pdf)).
235
 
236
  [UCI Machine Learning Repository - Credit Default Dataset](https://www.kaggle.com/datasets/uciml/default-of-credit-card-clients-dataset)
 
248
  - LIMIT_BAL signifies the amount of given credit in NT dollars (includes individual and family/supplementary credit).
249
  - BILL_AMT1 indicates the bill statement amount in September, 2005 (NT dollar).
250
 
251
+
252
+ HOME PRICE INTERPRETATION
253
+ =======================
254
+ This analysis is derived from an XGBoost regression model designed to predict house prices. The model utilizes features such as **dist_subway, age, lat, long,** and **dist_stores**.
255
+
256
+ Full dataset at the bottom of this tab
257
+
258
  Explain by Dataset
259
+ ----------
260
  - Below are explanation in typical background E[f(x)]
261
 
262
  ![Summary](file=./xgb/data.png)
 
269
 
270
 
271
  Explain by Feature
272
+ ----------
273
  ![Partial Dependence](file=./xgb/feature.png)
274
 
275
  **Observations:**
 
278
 
279
 
280
  Explain by Record
281
+ ----------
282
  ![Force](file=./xgb/record.png)
283
 
284
  **Contribution to Price:**
 
286
  - **Age** follows as the second significant contributor.
287
 
288
  Explain by Instance
289
+ ----------
290
  ![Dependence](file=./xgb/instance.png)
291
 
292
  **Insights:**