Update README.md

README.md

@@ -65,20 +65,23 @@ For a detailed example of how to load and use this model, please refer to the Co
 ## Ethical Considerations
 
 As with any language model, care should be taken when deploying this for real-world applications. Potential biases present in the training data could be reflected in the translations. It's important to monitor its output and ensure fair and accurate use.
-## 🚀 How to use
-
-```python
+# ==============================
+# LinguaFlow – English to Arabic Translator
+# ==============================
 from huggingface_hub import snapshot_download
 import tensorflow as tf
-import numpy as np
-import os
 from tensorflow.keras.preprocessing.text import tokenizer_from_json
 from tensorflow.keras.preprocessing.sequence import pad_sequences
+import os
+import numpy as np
 
-repo_id = "{repo_id}"
-local_dir = snapshot_download(repo_id=repo_id)
+# ------------------------------
+# ------------------------------
+repo_id = "Ali0044/LinguaFlow"
+local_dir = snapshot_download(repo_id=repo_id)  # لو الريبو خاص، استخدم token="HF_TOKEN"
 
-model = tf.keras.models.load_model(os.path.join(local_dir, "Translation_model_for_hf.keras"))
+model_path = os.path.join(local_dir, "Translation_model_for_hf.keras")
+model = tf.keras.models.load_model(model_path)
 
 with open(os.path.join(local_dir, "tokenizer/eng_tokenizer.json"), "r", encoding="utf-8") as f:
     eng_tokenizer = tokenizer_from_json(f.read())
@@ -86,18 +89,32 @@ with open(os.path.join(local_dir, "tokenizer/eng_tokenizer.json"), "r", encoding
 with open(os.path.join(local_dir, "tokenizer/ar_tokenizer.json"), "r", encoding="utf-8") as f:
     ar_tokenizer = tokenizer_from_json(f.read())
 
+# ------------------------------
+# ------------------------------
+def encode_text(texts, tokenizer, max_len):
+    seq = tokenizer.texts_to_sequences(texts)
+    return pad_sequences(seq, maxlen=max_len, padding='post')
+
+def sequences_to_text(sequences, tokenizer):
+    texts = []
+    for seq in sequences:
+        temp = [tokenizer.index_word[idx] for idx in seq if idx != 0]
+        texts.append(' '.join(temp))
+    return texts
+
+# ------------------------------
+# ------------------------------
 def translate(sentences):
-    seq = eng_tokenizer.texts_to_sequences(sentences)
-    padded = pad_sequences(seq, maxlen=model.input_shape[1], padding='post')
-    preds = model.predict(padded)
+    X_input = encode_text(sentences, eng_tokenizer, model.input_shape[1])
+    preds = model.predict(X_input)
     preds = np.argmax(preds, axis=-1)
-    
-    results = []
-    for s in preds:
-        text = [ar_tokenizer.index_word[i] for i in s if i != 0]
-        results.append(' '.join(text))
-    return results
-
-# Example
-print(translate(["Hello, how are you?"]))
-"""
+    return sequences_to_text(preds, ar_tokenizer)
+
+# ------------------------------
+# ------------------------------
+sample_texts = ["Hello, how are you?", "I love machine learning!"]
+translations = translate(sample_texts)
+
+for en, ar in zip(sample_texts, translations):
+    print(f"English: {en}")
+    print(f"Arabic : {ar}")