RICHARDMENSAH commited on
Commit
9ebc0c2
1 Parent(s): ab7726e
Files changed (1) hide show
  1. main.py +84 -0
main.py ADDED
@@ -0,0 +1,84 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ import pandas as pd
2
+ import joblib
3
+ from fastapi import FastAPI
4
+ import uvicorn
5
+ import numpy as np
6
+ import os
7
+
8
+ app = FastAPI()
9
+ #app.mount("/static", StaticFiles(directory="static"), name="static")
10
+ #templates = Jinja2Templates(directory="templates")
11
+
12
+ def load_model():
13
+ cwd = os.getcwd()
14
+ destination = os.path.join(cwd, "Assets")
15
+
16
+ imputer_filepath = os.path.join(destination, "numerical_imputer.joblib")
17
+ scaler_filepath = os.path.join(destination, "scaler.joblib")
18
+ model_filepath = os.path.join(destination, "lr_model.joblib")
19
+
20
+ num_imputer = joblib.load(imputer_filepath)
21
+ scaler = joblib.load(scaler_filepath)
22
+ model = joblib.load(model_filepath)
23
+
24
+ return num_imputer, scaler, model
25
+
26
+
27
+ def preprocess_input_data(input_data, num_imputer, scaler):
28
+ input_data_df = pd.DataFrame([input_data])
29
+ num_columns = [col for col in input_data_df.columns if input_data_df[col].dtype != 'object']
30
+ input_data_imputed_num = num_imputer.transform(input_data_df[num_columns])
31
+ input_scaled_df = pd.DataFrame(scaler.transform(input_data_imputed_num), columns=num_columns)
32
+ return input_scaled_df
33
+
34
+ @app.get("/")
35
+ def read_root():
36
+ return "Sepsis Prediction App"
37
+
38
+ @app.get("/sepsis/predict")
39
+ def predict_sepsis_endpoint(PRG: float, PL: float, PR: float, SK: float, TS: float,
40
+ M11: float, BD2: float, Age: float, Insurance: int):
41
+ num_imputer, scaler, model = load_model()
42
+
43
+ input_data = {
44
+ 'PRG': [PRG],
45
+ 'PL': [PL],
46
+ 'PR': [PR],
47
+ 'SK': [SK],
48
+ 'TS': [TS],
49
+ 'M11': [M11],
50
+ 'BD2': [BD2],
51
+ 'Age': [Age],
52
+ 'Insurance': [Insurance]
53
+ }
54
+
55
+ input_scaled_df = preprocess_input_data(input_data, num_imputer, scaler)
56
+
57
+ probabilities = model.predict_proba(input_scaled_df)[0]
58
+ prediction = np.argmax(probabilities)
59
+
60
+ sepsis_status = "Positive" if prediction == 1 else "Negative"
61
+
62
+ probability = probabilities[1] if prediction == 1 else probabilities[0]
63
+
64
+ #statement = f"The patient is {sepsis_status}. There is a {'high' if prediction == 1 else 'low'} probability ({probability:.2f}) that the patient is susceptible to developing sepsis."
65
+
66
+ if prediction == 1:
67
+ status_icon = "✔" # Red 'X' icon for positive sepsis prediction
68
+ sepsis_explanation = "Sepsis is a life-threatening condition caused by an infection. A positive prediction suggests that the patient might be exhibiting sepsis symptoms and requires immediate medical attention."
69
+ else:
70
+ status_icon = "✘" # Green checkmark icon for negative sepsis prediction
71
+ sepsis_explanation = "Sepsis is a life-threatening condition caused by an infection. A negative prediction suggests that the patient is not currently exhibiting sepsis symptoms."
72
+
73
+ statement = f"The patient's sepsis status is {sepsis_status} {status_icon} with a probability of {probability:.2f}. {sepsis_explanation}"
74
+
75
+ user_input_statement = "Please note this is the user-inputted data: "
76
+
77
+ output_df = pd.DataFrame([input_data])
78
+
79
+ result = {'predicted_sepsis': sepsis_status, 'statement': statement, 'user_input_statement': user_input_statement, 'input_data_df': output_df.to_dict('records')}
80
+
81
+ return result
82
+
83
+ if __name__ == "__main__":
84
+ uvicorn.run(app, host="0.0.0.0", port=7860, reload=True)