Q4. load diffrent model and run predict

api/app.py

@@ -1,9 +1,28 @@
 from flask import Flask, request, jsonify
 from joblib import load
 import os
+from markupsafe import escape
+
 
 app = Flask(__name__)
 
+def load_model(model_name):
+ dirname = os.path.dirname(__file__)
+ filename_svm = os.path.join(dirname, '../models/svmgamma:0.001_C:1.joblib')
+ svm = load(filename_svm)
+ filename_tree = os.path.join(dirname, '../models/treemax_depth:100.joblib')
+ tree = load(filename_tree)
+ filename_lr = os.path.join(dirname, '../models/m23csa018_lr_lbfgs.joblib')
+ lr = load(filename_lr)
+ if model_name == 'svm':
+ return svm
+ elif model_name == 'tree':
+ return tree
+ elif model_name == 'lr':
+ return lr
+ else:
+ return None
+
 @app.route('/')
 def index():
  return "Hello, World"
@@ -12,14 +31,13 @@ def index():
 def hello_world_post(): 
  return {"op" : "Hello, World POST " + request.json["suffix"]}
 
-@app.route('/predict', methods=['POST'])
+@app.route('/predict/<model_name>', methods=['POST'])
 def pred_model():
  js = request.get_json()
+ model_name = escape(model_name)
  image1 = [js['image']]
  #Assuming this is the path of our best trained model
- dirname = os.path.dirname(__file__)
- filename = os.path.join(dirname, '../models/treemax_depth:100.joblib')
- model = load(filename)
+ model = load_model(model_name)
  pred1 = model.predict(image1)
  #reurn pred1 in json
  return jsonify(prediction=pred1.tolist())