Spaces:
Running
Running
Update app.py
Browse files
app.py
CHANGED
@@ -70,12 +70,14 @@ else:
|
|
70 |
TITLE = "Multi-Tagger"
|
71 |
DESCRIPTION = """
|
72 |
Multi-Tagger is a powerful and versatile application that integrates two cutting-edge models: Waifu Diffusion and Florence 2. This app is designed to provide comprehensive image analysis and captioning capabilities, making it a valuable tool for AI artists, researchers, and enthusiasts.
|
|
|
73 |
Features:
|
74 |
- Supports batch processing of multiple images.
|
75 |
- Tags images with multiple categories: general tags, character tags, and ratings.
|
76 |
- Displays categorized tags in a structured format.
|
77 |
- Includes a separate tab for image captioning using Florence 2. This model supports CUDA, MPS or CPU if one of them is available.
|
78 |
- Supports various captioning tasks (e.g., Caption, Detailed Caption, Object Detection), as well it can display output text and images for tasks that generate visual outputs.
|
|
|
79 |
Example image by [me.](https://huggingface.co/Werli)
|
80 |
"""
|
81 |
colormap = ['blue','orange','green','purple','brown','pink','gray','olive','cyan','red',
|
@@ -213,6 +215,7 @@ class Llama3Reorganize:
|
|
213 |
loadModel: bool = False,
|
214 |
):
|
215 |
"""Initializes the Llama model.
|
|
|
216 |
Args:
|
217 |
repoId: LLAMA model repo.
|
218 |
device: Device to use for computation (cpu, cuda, ipu, xpu, mkldnn, opengl, opencl,
|
@@ -1020,7 +1023,6 @@ with gr.Blocks(title=TITLE, css=css, theme="Werli/Multi-Tagger", fill_width=True
|
|
1020 |
[
|
1021 |
download_file,
|
1022 |
sorted_general_strings,
|
1023 |
-
categorized_output,
|
1024 |
categorized,
|
1025 |
rating,
|
1026 |
character_res,
|
@@ -1035,7 +1037,7 @@ with gr.Blocks(title=TITLE, css=css, theme="Werli/Multi-Tagger", fill_width=True
|
|
1035 |
upload_button.upload(extend_gallery, inputs=[gallery, upload_button], outputs=gallery)
|
1036 |
# Event to update the selected image when an image is clicked in the gallery
|
1037 |
selected_image = gr.Textbox(label="Selected Image", visible=False)
|
1038 |
-
gallery.select(get_selection_from_gallery, inputs=[gallery, tag_results], outputs=[selected_image, sorted_general_strings, categorized,
|
1039 |
# Event to remove a selected image from the gallery
|
1040 |
remove_button.click(remove_image_from_gallery, inputs=[gallery, selected_image], outputs=gallery)
|
1041 |
submit.click(
|
@@ -1053,7 +1055,7 @@ with gr.Blocks(title=TITLE, css=css, theme="Werli/Multi-Tagger", fill_width=True
|
|
1053 |
additional_tags_append,
|
1054 |
tag_results,
|
1055 |
],
|
1056 |
-
outputs=[download_file, sorted_general_strings, categorized,
|
1057 |
)
|
1058 |
gr.Examples(
|
1059 |
[["images/1girl.png", VIT_LARGE_MODEL_DSV3_REPO, 0.35, False, 0.85, False]],
|
|
|
70 |
TITLE = "Multi-Tagger"
|
71 |
DESCRIPTION = """
|
72 |
Multi-Tagger is a powerful and versatile application that integrates two cutting-edge models: Waifu Diffusion and Florence 2. This app is designed to provide comprehensive image analysis and captioning capabilities, making it a valuable tool for AI artists, researchers, and enthusiasts.
|
73 |
+
|
74 |
Features:
|
75 |
- Supports batch processing of multiple images.
|
76 |
- Tags images with multiple categories: general tags, character tags, and ratings.
|
77 |
- Displays categorized tags in a structured format.
|
78 |
- Includes a separate tab for image captioning using Florence 2. This model supports CUDA, MPS or CPU if one of them is available.
|
79 |
- Supports various captioning tasks (e.g., Caption, Detailed Caption, Object Detection), as well it can display output text and images for tasks that generate visual outputs.
|
80 |
+
|
81 |
Example image by [me.](https://huggingface.co/Werli)
|
82 |
"""
|
83 |
colormap = ['blue','orange','green','purple','brown','pink','gray','olive','cyan','red',
|
|
|
215 |
loadModel: bool = False,
|
216 |
):
|
217 |
"""Initializes the Llama model.
|
218 |
+
|
219 |
Args:
|
220 |
repoId: LLAMA model repo.
|
221 |
device: Device to use for computation (cpu, cuda, ipu, xpu, mkldnn, opengl, opencl,
|
|
|
1023 |
[
|
1024 |
download_file,
|
1025 |
sorted_general_strings,
|
|
|
1026 |
categorized,
|
1027 |
rating,
|
1028 |
character_res,
|
|
|
1037 |
upload_button.upload(extend_gallery, inputs=[gallery, upload_button], outputs=gallery)
|
1038 |
# Event to update the selected image when an image is clicked in the gallery
|
1039 |
selected_image = gr.Textbox(label="Selected Image", visible=False)
|
1040 |
+
gallery.select(get_selection_from_gallery, inputs=[gallery, tag_results], outputs=[selected_image, sorted_general_strings, categorized, rating, character_res, general_res, unclassified])
|
1041 |
# Event to remove a selected image from the gallery
|
1042 |
remove_button.click(remove_image_from_gallery, inputs=[gallery, selected_image], outputs=gallery)
|
1043 |
submit.click(
|
|
|
1055 |
additional_tags_append,
|
1056 |
tag_results,
|
1057 |
],
|
1058 |
+
outputs=[download_file, sorted_general_strings, categorized, rating, character_res, general_res, unclassified, tag_results, categorized_output,],
|
1059 |
)
|
1060 |
gr.Examples(
|
1061 |
[["images/1girl.png", VIT_LARGE_MODEL_DSV3_REPO, 0.35, False, 0.85, False]],
|