radwaamr1 commited on
Commit
7ed4e27
·
verified ·
1 Parent(s): 1cab8d9

Upload 2 files

Browse files
Files changed (2) hide show
  1. app.py +73 -0
  2. requirements.txt +6 -0
app.py ADDED
@@ -0,0 +1,73 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # app.py
2
+ import gradio as gr
3
+ import pandas as pd
4
+ import joblib
5
+ import numpy as np
6
+ from sklearn.base import BaseEstimator, TransformerMixin
7
+
8
+ # -------------------------------
9
+ # Load saved artifacts
10
+ # -------------------------------
11
+ xgb_pipe = joblib.load("xgb_desertification_pipeline.pkl")
12
+ le = joblib.load("label_encoder.pkl")
13
+ expected_features = joblib.load("feature_names.pkl")
14
+
15
+ # -------------------------------
16
+ # FeatureFilter class
17
+ # -------------------------------
18
+ # Needed in case it's part of the pipeline
19
+ class FeatureFilter(BaseEstimator, TransformerMixin):
20
+ def __init__(self, expected_features):
21
+ self.expected_features = expected_features
22
+
23
+ def fit(self, X, y=None):
24
+ return self
25
+
26
+ def transform(self, X):
27
+ X = X.copy()
28
+ # Keep only expected features
29
+ X = X[[c for c in self.expected_features if c in X.columns]]
30
+ # Add missing columns with NaN
31
+ for c in self.expected_features:
32
+ if c not in X.columns:
33
+ X[c] = np.nan
34
+ return X[self.expected_features]
35
+
36
+ # -------------------------------
37
+ # Prediction function
38
+ # -------------------------------
39
+ def predict_desertification(**kwargs):
40
+ # Convert inputs to DataFrame
41
+ df = pd.DataFrame([kwargs])
42
+ df = df[expected_features] # reorder columns to match training
43
+ # Predict
44
+ pred_encoded = xgb_pipe.predict(df)
45
+ pred_label = le.inverse_transform(pred_encoded)
46
+ pred_proba = xgb_pipe.predict_proba(df).max(axis=1)
47
+ return {
48
+ "Desertification Level": pred_label[0],
49
+ "Confidence": float(pred_proba[0])
50
+ }
51
+
52
+ # -------------------------------
53
+ # Define Gradio interface
54
+ # -------------------------------
55
+ # Separate categorical and numeric inputs if needed
56
+ inputs = []
57
+ for feat in expected_features:
58
+ if feat == "lc_type1":
59
+ inputs.append(gr.Number(label=feat, value=0))
60
+ else:
61
+ inputs.append(gr.Number(label=feat, value=0.0))
62
+
63
+ # Gradio app
64
+ demo = gr.Interface(
65
+ fn=predict_desertification,
66
+ inputs=inputs,
67
+ outputs="json",
68
+ title="Desertification Level Predictor",
69
+ description="Predict desertification level given soil, climate, and land cover features."
70
+ )
71
+
72
+ # Launch
73
+ demo.launch()
requirements.txt ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ gradio
2
+ pandas
3
+ numpy
4
+ scikit-learn
5
+ xgboost
6
+ joblib