Creating test for image captioning using data from the database

backend/app/image_captioning/caption.py

@@ -0,0 +1,15 @@
+from PIL import Image
+from transformers import BlipProcessor, BlipForConditionalGeneration
+
+
+
+def caption1(img):
+    processor = BlipProcessor.from_pretrained("salesforce/blip-image-captioning-base")
+    model = BlipForConditionalGeneration.from_pretrained("salesforce/blip-image-captioning-base")
+
+
+    img_input = Image.fromarray(img)
+    inputs = processor(img_input, return_tensors="pt")
+    out = model.generate(**inputs)
+    caption = processor.decode(out[0], skip_special_token=True)
+    return caption

backend/app/image_captioning/test.py

@@ -0,0 +1,27 @@
+import os
+import sqlite3
+import numpy as np
+from PIL import Image
+from app.image_captioning.caption import caption1
+from app.config.settings import IMAGES_DATABASE_PATH
+
+
+def captioning_images():
+    conn = sqlite3.connect(IMAGES_DATABASE_PATH)
+    cursor = conn.cursor()
+
+    cursor.execute("SELECT path FROM image_id_mapping;")
+    paths = [row[0] for row in cursor.fetchall()]  
+
+    conn.close()
+    captions = []
+    for i in paths:
+        img = Image.open(os.path.abspath(i))  
+        img = np.array(img) 
+        caption = caption1(img)
+        print(f"{os.path.abspath(i)}:{caption}")
+        captions.append(caption)
+    print(captions)
+    print(paths)
+
+captioning_images()