Update README.md

README.md

@@ -45,34 +45,63 @@ The model achieved the following performance on the test set:
 
 ## 🚀 Usage Example
 
+Download model from HF
 ```python
+from huggingface_hub import hf_hub_download
 import pickle
-import numpy as np
 from tensorflow.keras.models import load_model
-from tensorflow.keras.preprocessing.sequence import pad_sequences
 
-# Load the model and tokenizer
-model = load_model("absa_bilstm_model.keras")
-with open("tokenizer.pkl", "rb") as f:
+model_path = hf_hub_download(repo_id="Sengil/Turkish-ABSA-BiLSTM-Word2Vec", filename="absa_bilstm_model.keras")
+tokenizer_path = hf_hub_download(repo_id="Sengil/Turkish-ABSA-BiLSTM-Word2Vec", filename="tokenizer.pkl")
+
+# load model
+model = load_model(model_path)
+
+# load tokenizer
+with open(tokenizer_path, "rb") as f:
     tokenizer = pickle.load(f)
+````
+
+Input preprocessing
+```python
+import re
+import nltk
+nltk.download('punkt')
+
+def preprocess_turkish(text):
+    text = text.lower()
+    text = re.sub(r"http\S+|www\S+|https\S+", "<url>", text)
+    text = re.sub(r"@\w+", "<user>", text)
+    text = re.sub(r"[^a-zA-Z0-9çğıöşüÇĞİÖŞÜ\s]", " ", text)
+    text = re.sub(r"(.)\1{2,}", r"\1\1", text)
+    text = re.sub(r"\s+", " ", text).strip()
+    return text
+````
 
-# Maximum sentence length used during training
-max_len = 84  # Adjust this value based on your training configuration
-
-# Prediction function
-def predict_sentiment(sentence, aspect):
-    input_text = f"{sentence} [ASP] {aspect}"
-    sequence = tokenizer.texts_to_sequences([input_text])
-    padded_sequence = pad_sequences(sequence, maxlen=max_len, padding='post')
-    prediction = model.predict(padded_sequence)
-    label = np.argmax(prediction, axis=1)[0]
-    labels = {0: "Negative", 1: "Neutral", 2: "Positive"}
+Predict the input
+```python
+import numpy as np
+from tensorflow.keras.preprocessing.sequence import pad_sequences
+
+def predict_sentiment(sentence, aspect, max_len=84):
+    input_text = sentence + " [ASP] " + aspect
+    cleaned = preprocess_turkish(input_text)
+    tokenized = tokenizer.texts_to_sequences([cleaned])
+    padded = pad_sequences(tokenized, maxlen=max_len, padding='post')
+    
+    pred = model.predict(padded)
+    label = np.argmax(pred)
+    labels = {0: "Negatif", 1: "Nötr", 2: "Pozitif"}
     return labels[label]
+````
+
+run
+```python
+sentence = "Manzara sahane evet ama servis rezalet."
+aspect = "manzara"
 
-# Example usage
-sentence = "Manzara şahane evet ama servis rezalet."
-aspect = "Servis"
-print(f"Sentiment for '{aspect}': {predict_sentiment(sentence, aspect)}")
+predict = predict_sentiment(sentence, aspect)
+print("predict:", predict)
 ````
 
 ## 🏋️‍♀️ Training Details