Upload 5 files

handler.py

@@ -1,56 +1,55 @@
-import nltk
-# Ensure NLTK data is downloaded
-nltk.download('punkt')
-
-import joblib
-import os
-import json
-
-
-
-class EndpointHandler:
-    def __init__(self, model_dir):
-        self.model_dir = model_dir
-        self.vectorizer = joblib.load(os.path.join(model_dir, 'vectorizer.joblib'))
-        self.model = joblib.load(os.path.join(model_dir, 'logistic_classifier.joblib'))
-
-        # Define the custom tokenizer function
-        self.tokenizer = nltk.word_tokenize
-
-        # Verify that the tokenizer configuration is correct
-        with open(os.path.join(model_dir, "tokenizer.json"), "r") as file:
-            tokenizer_config = json.load(file)
-        if tokenizer_config['tokenizer'] != 'nltk.word_tokenize':
-            raise ValueError("Tokenizer configuration does not match the expected tokenizer.")
-
-    def predict_rating(self, review):
-        review_tfidf = self.vectorizer.transform([review])
-        predicted_rating = self.model.predict(review_tfidf)[0]
-        return int(predicted_rating)
-
-    def __call__(self, inputs):
-        try:
-            inputs_dict = json.loads(inputs)
-
-            if 'review' not in inputs_dict:
-                return json.dumps({"error": "No valid review provided"})
-
-            review = inputs_dict['review']
-
-            if not review:
-                return json.dumps({"error": "No valid review provided"})
-
-            predicted_rating = self.predict_rating(review)
-
-            response = {
-                "review": review,
-                "predicted_rating": predicted_rating
-            }
-
-            return json.dumps(response)
-
-        except json.JSONDecodeError:
-            return json.dumps({"error": "Invalid JSON format"})
-
-        except Exception as e:
-            return json.dumps({"error": str(e)})
+import joblib
+import os
+import json
+
+class EndpointHandler:
+    def __init__(self, model_dir):
+        self.model_dir = model_dir
+        self.vectorizer = joblib.load(os.path.join(model_dir, 'vectorizer.joblib'))
+        self.model = joblib.load(os.path.join(model_dir, 'logistic_classifier.joblib'))
+
+        # Define the custom tokenizer function
+        def custom_tokenizer(text):
+            return text.split()
+        
+        self.tokenizer = custom_tokenizer
+
+        # Verify that the tokenizer configuration is correct
+        with open(os.path.join(model_dir, "tokenizer.json"), "r") as file:
+            tokenizer_config = json.load(file)
+        if tokenizer_config['tokenizer'] != 'split':
+            raise ValueError("Tokenizer configuration does not match the expected tokenizer.")
+
+    def predict_rating(self, review):
+        review_tfidf = self.vectorizer.transform([review])
+        predicted_rating = self.model.predict(review_tfidf)[0]
+        return int(predicted_rating)
+
+    def __call__(self, inputs):
+        try:
+            inputs_dict = json.loads(inputs)
+
+            if 'review' not in inputs_dict:
+                return json.dumps({"error": "No valid review provided"})
+
+            review = inputs_dict['review']
+
+            if not review:
+                return json.dumps({"error": "No valid review provided"})
+
+            predicted_rating = self.predict_rating(review)
+
+            response = {
+                "review": review,
+                "predicted_rating": predicted_rating
+            }
+
+            return json.dumps(response)
+
+        except json.JSONDecodeError:
+            return json.dumps({"error": "Invalid JSON format"})
+
+        except Exception as e:
+            return json.dumps({"error": str(e)})
+
+

logistic_classifier.joblib

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:66ca534a1485c744bb11cfb827f4af94e1e8d46182267d5e2fe0d33b67d8df3e
+oid sha256:1320450110eb5efffddbb80edd4c1ffbad72753b2529db2f3ab79aac637883dc
 size 200927

requirement.txt

@@ -1,5 +1,4 @@
-joblib
-scikit-learn
-pandas
-nlp
-nltk
+joblib
+scikit-learn
+pandas
+nlp

tokenizer.json

@@ -1 +1 @@
-{"tokenizer": "nltk.word_tokenize"}
+{"tokenizer": "split"}

vectorizer.joblib

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:32f170a717c19ff2a6fc4480f609d6fd3252669cc88a1b4e38842d6dbdb7bd4a
-size 161885
+oid sha256:3533c6722416bede62ff5d61e33be74a6f6691fc077c228a5edbd78ed018d1db
+size 162076