Upload 15 files

.gitattributes

@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+data/creditcard.csv filter=lfs diff=lfs merge=lfs -text
+data/training.csv filter=lfs diff=lfs merge=lfs -text

data/creditcard.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:76274b691b16a6c49d3f159c883398e03ccd6d1ee12d9d8ee38f4b4b98551a89
+size 150828752

data/training.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:08cba6317a49528fcd074f9043aafcd5ad6c6be45ede159c4e36cec33af24afe
+size 238803811

src/data/preprocess.py

@@ -0,0 +1,19 @@
+import re
+from nltk.corpus import stopwords
+from nltk.stem.porter import PorterStemmer
+
+port_stem = PorterStemmer()
+stop_words = set(stopwords.words('english'))
+
+def clean_text(text: str) -> str:
+    text = re.sub('[^a-zA-Z]', ' ', str(text))
+    words = text.lower().split()
+
+    words = [
+        port_stem.stem(word)
+        for word in words
+        if word not in stop_words
+    ]
+
+    return ' '.join(words)
+

src/model/infernce.py

@@ -0,0 +1,12 @@
+import pickle
+import pandas as pd
+
+class SentimentModel:
+
+    def __init__(self, model_path='artifacts/model.pkl'):
+        with open(model_path, 'rb') as f:
+            self.model = pickle.load(f)
+
+    def predict(self, text: str):
+        pred = self.model.predict([text])[0]
+        return "Positive" if pred == 1 else "Negative"

src/model/train.py

@@ -0,0 +1,46 @@
+import pandas as pd
+import pickle
+from sklearn.model_selection import train_test_split
+from sklearn.metrics import accuracy_score
+
+from src.pipeline.build_pipeline import build_pipeline
+
+
+def train_model():
+    columns = ['target', 'id', 'date','flag','user','text']
+
+    data = pd.read_csv(
+    'data/training.csv',
+    names=columns,
+    encoding='ISO-8859-1'
+)
+
+    data.replace({'target': {4: 1}}, inplace=True)
+
+    
+    data = data.sample(1000, random_state=42)
+
+    X = data['text']
+    y = data['target']
+
+    x_train, x_test, y_train, y_test = train_test_split(
+        X, y, test_size=0.2, stratify=y, random_state=2
+    )
+
+    pipeline = build_pipeline()
+
+    pipeline.fit(x_train, y_train)
+
+    print("Train Accuracy:", accuracy_score(y_train, pipeline.predict(x_train)))
+    print("Test Accuracy:", accuracy_score(y_test, pipeline.predict(x_test)))
+
+    
+    with open('artifacts/model.pkl', 'wb') as f:
+        pickle.dump(pipeline, f)
+
+if __name__ == "__main__":
+    train_model()
+    
+
+
+

src/models/inference.py

@@ -0,0 +1,12 @@
+import pickle
+import pandas as pd
+
+class SentimentModel:
+
+    def __init__(self, model_path='artifacts/model.pkl'):
+        with open(model_path, 'rb') as f:
+            self.model = pickle.load(f)
+
+    def predict(self, text: str):
+        pred = self.model.predict([text])[0]
+        return "Positive" if pred == 1 else "Negative"

src/models/train.py

@@ -0,0 +1,45 @@
+import pandas as pd
+import pickle
+from sklearn.model_selection import train_test_split
+from sklearn.metrics import accuracy_score
+
+from src.pipeline.build_pipeline import build_pipeline
+
+
+def train_model():
+    columns = ['target', 'id', 'date','flag','user','text']
+
+    data = pd.read_csv(
+    'data/training.csv',
+    names=columns,
+    encoding='ISO-8859-1'
+)
+
+    data.replace({'target': {4: 1}}, inplace=True)
+
+    
+    data = data.sample(1000, random_state=42)
+
+    X = data['text']
+    y = data['target']
+
+    x_train, x_test, y_train, y_test = train_test_split(
+        X, y, test_size=0.2, stratify=y, random_state=2
+    )
+
+    pipeline = build_pipeline()
+
+    pipeline.fit(x_train, y_train)
+
+    print("Train Accuracy:", accuracy_score(y_train, pipeline.predict(x_train)))
+    print("Test Accuracy:", accuracy_score(y_test, pipeline.predict(x_test)))
+
+    
+    with open('artifacts/model.pkl', 'wb') as f:
+        pickle.dump(pipeline, f)
+# testing
+# def test_ok():
+#     hdshsdhfs
+if __name__ == "__main__":
+    train_model()
+    

src/pipeline/build_pipeline.py

@@ -0,0 +1,15 @@
+from sklearn.pipeline import Pipeline
+from sklearn.feature_extraction.text import TfidfVectorizer
+from sklearn.linear_model import LogisticRegression
+
+from src.data.preprocess import clean_text
+
+
+
+def build_pipeline():
+    pipeline = Pipeline([
+        ('tfidf', TfidfVectorizer(preprocessor=clean_text)),
+        ('model', LogisticRegression(max_iter=1000))
+    ])
+
+    return pipeline