cleaned code and set output for amputation risk
Browse files
app.py
CHANGED
@@ -17,42 +17,23 @@ def predict(model, input_df):
|
|
17 |
predictions_df = predict_model(estimator=model, data=input_df)
|
18 |
predict_label = predictions_df["Label"][0] # either 1 (amputation yes) or 0 (amputation no)
|
19 |
predict_score = predictions_df["Score"][0] # the prediction (accuracy)
|
20 |
-
|
21 |
-
|
22 |
-
|
23 |
-
|
24 |
-
|
25 |
-
|
26 |
-
|
27 |
-
|
28 |
-
|
29 |
-
|
30 |
|
31 |
|
32 |
# the parameters in this function, actually gets the inputs for the prediction
|
33 |
def predict_amputation(age, gender, race, diabetes_type):
|
34 |
-
#input_dict = {"AGE": 70.0, "GENDER_F": 0.0, "RACE_Asian": 1.0, "RACE_Black": 0.0, "RACE_Coloured": 0.0, "RACE_Other": 0.0, "RACE_White": 0.0, "DIABETES_CLASS_Type 1 diabetes":0.0}
|
35 |
-
|
36 |
-
|
37 |
-
#input_dict = {"AGE": 70.0, "GENDER": 0.0, "RACE": 1.0, "DIABETES_CLASS":0.0, "AMPUTATION":0}
|
38 |
-
|
39 |
-
|
40 |
-
#input_dict = {"AGE": 70, "GENDER": "F", "RACE": "Asian", "DIABETES_CLASS":"Type 2 diabetes", "AMPUTATION":''}
|
41 |
-
|
42 |
-
|
43 |
-
#input_dict = {"AGE": 80, "GENDER": "F", "RACE": "Asian", "DIABETES_CLASS":"Type 2 diabetes", "AMPUTATION":''}
|
44 |
-
|
45 |
diabetes_class = "Type "+str(diabetes_type)+" diabetes"
|
46 |
gender = gender[0]
|
47 |
input_dict = {"AGE": age, "GENDER": gender, "RACE": race, "DIABETES_CLASS":diabetes_class, "AMPUTATION":''}
|
48 |
|
49 |
input_df = pd.DataFrame([input_dict])
|
50 |
|
51 |
-
return
|
52 |
-
#return str("ALLAH " + " " + str(age) + " " + gender + " " + race + diabetes_type)
|
53 |
-
#return diabetes_type
|
54 |
-
#return "ALLAH: "+str(predict(model=model, input_df=input_df)) # calls the predict function when 'submit' is clicked
|
55 |
-
|
56 |
|
57 |
|
58 |
|
@@ -63,8 +44,6 @@ description = "A diabetes-related amputation machine learning model trained on t
|
|
63 |
article = "<p style='text-align: center'><span style='font-size: 15pt;'>Copyright © DIARC. 2021. All Rights Reserved. Contact Us: <a href='mailto:smtshali@wol.co.za'>Dr Sifisiso Mtshali</a> or <a href='mailto:mahomedo@ukzn.ac.za'>Dr Ozayr Mahomed</a></span></p>"
|
64 |
|
65 |
|
66 |
-
|
67 |
-
|
68 |
iface = gr.Interface(
|
69 |
fn=predict_amputation,
|
70 |
title=title,
|
@@ -82,6 +61,7 @@ iface = gr.Interface(
|
|
82 |
],
|
83 |
)
|
84 |
|
|
|
85 |
iface.test_launch()
|
86 |
if __name__ == "__main__":
|
87 |
iface.launch()
|
|
|
17 |
predictions_df = predict_model(estimator=model, data=input_df)
|
18 |
predict_label = predictions_df["Label"][0] # either 1 (amputation yes) or 0 (amputation no)
|
19 |
predict_score = predictions_df["Score"][0] # the prediction (accuracy)
|
20 |
+
amputation_risk = ""
|
21 |
+
if predict_label == 1:
|
22 |
+
amputation_risk = "YES"
|
23 |
+
else:
|
24 |
+
amputation_risk = "NO"
|
25 |
+
return "AMPUTATION RISK: " + amputation_risk + " SCORE: "+str(predict_score)
|
|
|
|
|
|
|
|
|
26 |
|
27 |
|
28 |
# the parameters in this function, actually gets the inputs for the prediction
|
29 |
def predict_amputation(age, gender, race, diabetes_type):
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
30 |
diabetes_class = "Type "+str(diabetes_type)+" diabetes"
|
31 |
gender = gender[0]
|
32 |
input_dict = {"AGE": age, "GENDER": gender, "RACE": race, "DIABETES_CLASS":diabetes_class, "AMPUTATION":''}
|
33 |
|
34 |
input_df = pd.DataFrame([input_dict])
|
35 |
|
36 |
+
return str(predict(model=model, input_df=input_df)) # calls the predict function when 'submit' is clicked
|
|
|
|
|
|
|
|
|
37 |
|
38 |
|
39 |
|
|
|
44 |
article = "<p style='text-align: center'><span style='font-size: 15pt;'>Copyright © DIARC. 2021. All Rights Reserved. Contact Us: <a href='mailto:smtshali@wol.co.za'>Dr Sifisiso Mtshali</a> or <a href='mailto:mahomedo@ukzn.ac.za'>Dr Ozayr Mahomed</a></span></p>"
|
45 |
|
46 |
|
|
|
|
|
47 |
iface = gr.Interface(
|
48 |
fn=predict_amputation,
|
49 |
title=title,
|
|
|
61 |
],
|
62 |
)
|
63 |
|
64 |
+
|
65 |
iface.test_launch()
|
66 |
if __name__ == "__main__":
|
67 |
iface.launch()
|