Pls work

main.py

@@ -20,14 +20,19 @@ text = "a cat sitting on a table"
 # Tokenize text and get features
 inputs = clip_processor(text, return_tensors="pt", padding=True)
 
-# Generate image features
-image_features = clip_model.get_image_features(pixel_values=inputs.pixel_values)
+# Generate image from text
+generated_image = clip_model.generate_images(
+  input_ids=inputs.input_ids,
+  attention_mask=inputs.attention_mask,
+  visual_input=None,  # We don't provide image input
+  return_tensors="pt"  # Return PyTorch tensor
+)
 
-# Generate image from features
-generated_image = clip_model.generate_images(image_features)
+# Convert the generated image tensor to a NumPy array
+generated_image_np = generated_image[0].cpu().numpy()
 
 # Save the generated image
 output_image_path = "generated_image.png"
-Image.fromarray(generated_image).save(output_image_path)
+Image.fromarray(generated_image_np).save(output_image_path)
 
 print("Image generated and saved as:", output_image_path)