Create pipeline.py

pipeline.py

@@ -0,0 +1,33 @@
+import numpy as np
+from tensorflow.keras.preprocessing.sequence import pad_sequences
+from tensorflow.keras.preprocessing.text import tokenizer_from_json
+import tensorflow as tf
+import json
+import os
+
+class ToxicPipeline:
+    def __init__(self, model, tokenizer_path="tokenizer.json", max_len=150, label_map=None):
+        self.model = model
+        with open(tokenizer_path, "r", encoding="utf-8") as f:
+            tokenizer_json = f.read()
+            self.tokenizer = tokenizer_from_json(tokenizer_json)
+        self.max_len = max_len
+        self.label_map = label_map
+
+    def __call__(self, text, image_desc):
+        input_text = text + " " + image_desc
+        seq = self.tokenizer.texts_to_sequences([input_text])
+        padded = pad_sequences(seq, maxlen=self.max_len, padding='post', truncating='post')
+        pred_probs = self.model.predict(padded)
+        pred_label = int(np.argmax(pred_probs, axis=1)[0])
+        if self.label_map:
+            return self.label_map.get(pred_label, pred_label)
+        return pred_label
+
+# Example usage (for README):
+# from huggingface_hub import from_pretrained_keras
+# from pipeline import ToxicPipeline
+# model = from_pretrained_keras("NightPrince/Toxic_Classification")
+# pipeline = ToxicPipeline(model, tokenizer_path="data/tokenizer.json", label_map={0: "toxic", 1: "not toxic", ...})
+# result = pipeline("This is a dangerous post", "Knife shown in the image")
+# print(result)