DongfuJiang
commited on
Commit
•
75c15ae
1
Parent(s):
3f7f343
update
Browse files- app.py +1 -1
- models/mllava/utils.py +1 -1
app.py
CHANGED
@@ -109,7 +109,7 @@ def build_demo():
|
|
109 |
gr.Markdown(""" # Mantis
|
110 |
Mantis is a multimodal conversational AI model that can chat with users about images and text. It's optimized for multi-image reasoning, where inverleaved text and images can be used to generate responses.
|
111 |
|
112 |
-
### [Paper](https://arxiv.org/abs/2405.01483) | [Github](https://github.com/TIGER-AI-Lab/Mantis) | [Models](https://huggingface.co/collections/TIGER-Lab/mantis-6619b0834594c878cdb1d6e4) | [Dataset](https://huggingface.co/datasets/TIGER-Lab/Mantis-Instruct)
|
113 |
""")
|
114 |
|
115 |
gr.Markdown("""## Chat with Mantis
|
|
|
109 |
gr.Markdown(""" # Mantis
|
110 |
Mantis is a multimodal conversational AI model that can chat with users about images and text. It's optimized for multi-image reasoning, where inverleaved text and images can be used to generate responses.
|
111 |
|
112 |
+
### [Paper](https://arxiv.org/abs/2405.01483) | [Github](https://github.com/TIGER-AI-Lab/Mantis) | [Models](https://huggingface.co/collections/TIGER-Lab/mantis-6619b0834594c878cdb1d6e4) | [Dataset](https://huggingface.co/datasets/TIGER-Lab/Mantis-Instruct) | [Website](https://tiger-ai-lab.github.io/Mantis/)
|
113 |
""")
|
114 |
|
115 |
gr.Markdown("""## Chat with Mantis
|
models/mllava/utils.py
CHANGED
@@ -55,7 +55,7 @@ def chat_mllava(
|
|
55 |
if images:
|
56 |
for i in range(len(images)):
|
57 |
if isinstance(images[i], str):
|
58 |
-
images[i] = PIL.Image.open(images[i])
|
59 |
|
60 |
inputs = processor(images=images, text=prompt, return_tensors="pt", truncation=True, max_length=max_input_length)
|
61 |
for k, v in inputs.items():
|
|
|
55 |
if images:
|
56 |
for i in range(len(images)):
|
57 |
if isinstance(images[i], str):
|
58 |
+
images[i] = PIL.Image.open(images[i]).convert("RGB")
|
59 |
|
60 |
inputs = processor(images=images, text=prompt, return_tensors="pt", truncation=True, max_length=max_input_length)
|
61 |
for k, v in inputs.items():
|