rakesh1248 commited on
Commit
4ab35d4
·
verified ·
1 Parent(s): 10e6bb8

Upload app.py with huggingface_hub

Browse files
Files changed (1) hide show
  1. app.py +22 -48
app.py CHANGED
@@ -1,20 +1,12 @@
1
-
2
  import streamlit as st
3
  import pandas as pd
4
  import joblib
5
  from huggingface_hub import snapshot_download
6
  import os
7
- import mlflow
8
- import shutil
9
-
10
- # Disable MLflow tracking to avoid creating new runs in the deployed app
11
- mlflow.set_tracking_uri("file:///dev/null")
12
 
13
- # Define Hugging Face details for model download
14
- hf_username = "rakesh1248"
15
- # Configuration for loading the model from Hugging Face Models
16
- model_repo_id_app = "rakesh1248/random_forest_engine_condition_classifier" # Use the full repo_id
17
- model_path_in_repo_app = "random_forest_model.joblib"
18
 
19
  model_dir_app = "./model_cache"
20
  os.makedirs(model_dir_app, exist_ok=True)
@@ -27,42 +19,30 @@ def load_model_app():
27
  local_dir=model_dir_app
28
  )
29
 
30
- # Construct path to model file
31
- model_path = os.path.join(repo_path, random_forest_model.joblib)
32
 
33
- # Load model directly using joblib
34
  model = joblib.load(model_path)
35
-
36
- # local_model_path = hf_hub_download(
37
- # repo_id=model_repo_id_app,
38
- # filename=model_path_in_repo_app,
39
- # repo_type="model",
40
- # local_dir=model_dir_app
41
- # )
42
  return model
 
43
  except Exception as e:
44
- st.error(f"Error loading model from Hugging Face Hub: {e}")
45
  st.stop()
46
 
47
  loaded_model_app = load_model_app()
48
 
49
- # Streamlit Application Layout
50
  st.set_page_config(layout="wide")
51
  st.title("Engine Predictive Maintenance App")
52
- st.write("Predict whether an engine requires maintenance based on sensor readings.")
53
 
54
- # Sidebar for user inputs
55
  st.sidebar.header("Engine Sensor Readings")
56
 
57
- # Input widgets for numerical features
58
- engine_rpm = st.sidebar.slider("Engine RPM", min_value=60, max_value=2300, value=750)
59
- lub_oil_pressure = st.sidebar.slider("Lub Oil Pressure (bar/kPa)", min_value=0.0, max_value=8.0, value=3.5, step=0.1)
60
- fuel_pressure = st.sidebar.slider("Fuel Pressure (bar/kPa)", min_value=0.0, max_value=22.0, value=6.0, step=0.1)
61
- coolant_pressure = st.sidebar.slider("Coolant Pressure (bar/kPa)", min_value=0.0, max_value=8.0, value=2.0, step=0.1)
62
- lub_oil_temp = st.sidebar.slider("Lub Oil Temperature (°C)", min_value=70.0, max_value=90.0, value=78.0, step=0.1)
63
- coolant_temp = st.sidebar.slider("Coolant Temperature (°C)", min_value=60.0, max_value=200.0, value=80.0, step=0.1)
64
 
65
- # Create a DataFrame from user inputs
66
  input_data = pd.DataFrame([{
67
  'Engine rpm': engine_rpm,
68
  'Lub oil pressure': lub_oil_pressure,
@@ -72,22 +52,16 @@ input_data = pd.DataFrame([{
72
  'Coolant temp': coolant_temp
73
  }])
74
 
75
- st.subheader("Current Input Parameters:")
76
  st.write(input_data)
77
 
78
- # Prediction button
79
- if st.button("Predict Engine Condition"):
80
- if loaded_model_app:
81
- prediction = loaded_model_app.predict(input_data)
82
- prediction_proba = loaded_model_app.predict_proba(input_data)
83
-
84
- st.subheader("Prediction Result:")
85
- if prediction[0] == 1:
86
- st.error("Engine Condition: Faulty (Maintenance Recommended)")
87
- else:
88
- st.success("Engine Condition: Normal")
89
 
90
- st.write(f"Confidence (Normal): {prediction_proba[0][0]:.2f}")
91
- st.write(f"Confidence (Faulty): {prediction_proba[0][1]:.2f}")
92
  else:
93
- st.warning("Model not loaded. Please check the logs for errors.")
 
 
 
 
 
1
  import streamlit as st
2
  import pandas as pd
3
  import joblib
4
  from huggingface_hub import snapshot_download
5
  import os
 
 
 
 
 
6
 
7
+ # --- Model Config --- #
8
+ model_repo_id_app = "rakesh1248/random_forest_engine_condition_classifier"
9
+ model_filename = "random_forest_model.joblib"
 
 
10
 
11
  model_dir_app = "./model_cache"
12
  os.makedirs(model_dir_app, exist_ok=True)
 
19
  local_dir=model_dir_app
20
  )
21
 
22
+ model_path = os.path.join(repo_path, model_filename)
 
23
 
 
24
  model = joblib.load(model_path)
 
 
 
 
 
 
 
25
  return model
26
+
27
  except Exception as e:
28
+ st.error(f"Error loading model: {e}")
29
  st.stop()
30
 
31
  loaded_model_app = load_model_app()
32
 
33
+ # --- UI --- #
34
  st.set_page_config(layout="wide")
35
  st.title("Engine Predictive Maintenance App")
 
36
 
 
37
  st.sidebar.header("Engine Sensor Readings")
38
 
39
+ engine_rpm = st.sidebar.slider("Engine RPM", 60, 2300, 750)
40
+ lub_oil_pressure = st.sidebar.slider("Lub Oil Pressure", 0.0, 8.0, 3.5, 0.1)
41
+ fuel_pressure = st.sidebar.slider("Fuel Pressure", 0.0, 22.0, 6.0, 0.1)
42
+ coolant_pressure = st.sidebar.slider("Coolant Pressure", 0.0, 8.0, 2.0, 0.1)
43
+ lub_oil_temp = st.sidebar.slider("Lub Oil Temperature", 70.0, 90.0, 78.0, 0.1)
44
+ coolant_temp = st.sidebar.slider("Coolant Temperature", 60.0, 200.0, 80.0, 0.1)
 
45
 
 
46
  input_data = pd.DataFrame([{
47
  'Engine rpm': engine_rpm,
48
  'Lub oil pressure': lub_oil_pressure,
 
52
  'Coolant temp': coolant_temp
53
  }])
54
 
 
55
  st.write(input_data)
56
 
57
+ if st.button("Predict"):
58
+ prediction = loaded_model_app.predict(input_data)
59
+ proba = loaded_model_app.predict_proba(input_data)
 
 
 
 
 
 
 
 
60
 
61
+ if prediction[0] == 1:
62
+ st.error("Faulty Engine")
63
  else:
64
+ st.success("Normal Engine")
65
+
66
+ st.write(f"Normal: {proba[0][0]:.2f}")
67
+ st.write(f"Faulty: {proba[0][1]:.2f}")