From a42026c182a1337d3eedfec9d53fa9aefbff8775 Mon Sep 17 00:00:00 2001
From: webspam0 <173999334+webspam0@users.noreply.github.com>
Date: Thu, 18 Jul 2024 15:29:49 +1000
Subject: [PATCH] Fix caption output workaround - use ComfyUI
---
nodes.py | 12 +++---------
1 file changed, 3 insertions(+), 9 deletions(-)
diff --git a/nodes.py b/nodes.py
index 21f77cf..1255b56 100644
--- a/nodes.py
+++ b/nodes.py
@@ -127,6 +127,7 @@ class Florence2Run:
}
}
+ OUTPUT_IS_LIST = (False, False, True,)
RETURN_TYPES = ("IMAGE", "MASK", "STRING",)
RETURN_NAMES =("image", "mask", "caption",)
FUNCTION = "encode"
@@ -199,11 +200,7 @@ class Florence2Run:
clean_results = clean_results.replace('', '')
clean_results = clean_results.replace('', '')
- #return single string if only one image for compatibility with nodes that can't handle string lists
- if len(image) == 1:
- out_results = clean_results
- else:
- out_results.append(clean_results)
+ out_results.append(clean_results)
W, H = image_pil.size
parsed_answer = processor.post_process_generation(results, task=task_prompt, image_size=(W, H))
@@ -394,10 +391,7 @@ class Florence2Run:
results = processor.batch_decode(generated_ids, skip_special_tokens=False)[0]
clean_results = results.replace('', '').replace('', '')
- if len(image) == 1:
- out_results = clean_results
- else:
- out_results.append(clean_results)
+ out_results.append(clean_results)
out.append(F.to_tensor(image_pil).unsqueeze(0).permute(0, 2, 3, 1).cpu().float())