PavanShivaKumar commited on
Commit
47791fc
·
verified ·
1 Parent(s): 6988e29

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +14 -20
app.py CHANGED
@@ -1,25 +1,19 @@
1
  import numpy as np
2
  import gradio as gr
3
  import pickle
 
4
 
5
- # # Load the trained model
6
- # def load_model():
7
- # with open("random_forest_model.pkl", "rb") as f:
8
- # model = pickle.load(f)
9
- # return model
10
-
11
-
12
- import joblib
13
-
14
- # Load the trained model (if saved using joblib)
15
  def load_model():
16
- model = joblib.load("random_forest_model.pkl")
 
17
  return model
18
 
19
  # Prediction function
20
  def predict_project(fname, lname, email, role, experience, skills):
21
- # Load the model
22
  model = load_model()
 
23
 
24
  # Prepare the new employee data
25
  new_employee = {
@@ -35,15 +29,15 @@ def predict_project(fname, lname, email, role, experience, skills):
35
  role_mapping = {'Intern': 0, 'Senior': 1} # Map 'Intern' -> 0, 'Senior' -> 1
36
  new_employee['role'] = role_mapping.get(new_employee['role'], -1) # Default to -1 if unknown role
37
 
38
- # If skills need to be transformed, you can handle that here (e.g., using a pre-trained vectorizer).
39
- # For simplicity, we assume skills are directly passed.
40
-
41
- # Combine the features (experience, role)
42
- X_new_employee = np.array([[new_employee['experience'], new_employee['role']]])
43
 
44
- # If you had a vectorizer for skills, you could do something like:
45
- # skills_vector = vectorizer.transform([new_employee['skills']])
46
- # X_new_employee = np.hstack((X_new_employee, skills_vector.toarray()))
 
 
 
47
 
48
  # Predict the project ID for the new employee
49
  predicted_project = model.predict(X_new_employee)[0]
 
1
  import numpy as np
2
  import gradio as gr
3
  import pickle
4
+ from sklearn.feature_extraction.text import TfidfVectorizer
5
 
6
+ # Load the trained model (same as before)
 
 
 
 
 
 
 
 
 
7
  def load_model():
8
+ with open("random_forest_model.pkl", "rb") as f:
9
+ model = pickle.load(f)
10
  return model
11
 
12
  # Prediction function
13
  def predict_project(fname, lname, email, role, experience, skills):
14
+ # Load the model and vectorizer
15
  model = load_model()
16
+ vectorizer=TfidfVectorizer()
17
 
18
  # Prepare the new employee data
19
  new_employee = {
 
29
  role_mapping = {'Intern': 0, 'Senior': 1} # Map 'Intern' -> 0, 'Senior' -> 1
30
  new_employee['role'] = role_mapping.get(new_employee['role'], -1) # Default to -1 if unknown role
31
 
32
+ # Transform the skills using the same vectorizer used during training
33
+ skills_vector = vectorizer.transform([new_employee['skills']])
 
 
 
34
 
35
+ # Combine the features (experience, role, and skills)
36
+ X_new_employee = np.hstack((
37
+ np.array([[new_employee['experience']]]), # Reshape experience to 2D
38
+ np.array([[new_employee['role']]]), # Reshape role to 2D
39
+ skills_vector.toarray() # Convert skills vector to array
40
+ ))
41
 
42
  # Predict the project ID for the new employee
43
  predicted_project = model.predict(X_new_employee)[0]