Spaces:
Running
on
Zero
Running
on
Zero
Update app.py
Browse files
app.py
CHANGED
@@ -285,9 +285,9 @@ def generate_video(model_name: str, text: str, video_path: str,
|
|
285 |
|
286 |
# Define examples for image and video inference
|
287 |
image_examples = [
|
288 |
-
["Convert this page to docling", "images/1.png"
|
289 |
-
["OCR the image", "images/2.jpg"
|
290 |
-
["Convert this page to docling", "images/3.png"
|
291 |
]
|
292 |
|
293 |
video_examples = [
|
@@ -307,7 +307,7 @@ css = """
|
|
307 |
|
308 |
# Create the Gradio Interface
|
309 |
with gr.Blocks(css=css, theme="bethecloud/storj_theme") as demo:
|
310 |
-
gr.Markdown("# **[
|
311 |
with gr.Row():
|
312 |
with gr.Column():
|
313 |
with gr.Tabs():
|
@@ -315,38 +315,32 @@ with gr.Blocks(css=css, theme="bethecloud/storj_theme") as demo:
|
|
315 |
image_query = gr.Textbox(label="Query Input", placeholder="Enter your query here...")
|
316 |
image_upload = gr.Image(type="pil", label="Image")
|
317 |
image_submit = gr.Button("Submit", elem_classes="submit-btn")
|
|
|
|
|
|
|
|
|
318 |
with gr.TabItem("Video Inference"):
|
319 |
video_query = gr.Textbox(label="Query Input", placeholder="Enter your query here...")
|
320 |
video_upload = gr.Video(label="Video")
|
321 |
video_submit = gr.Button("Submit", elem_classes="submit-btn")
|
|
|
|
|
|
|
|
|
322 |
with gr.Accordion("Advanced options", open=False):
|
323 |
max_new_tokens = gr.Slider(label="Max new tokens", minimum=1, maximum=MAX_MAX_NEW_TOKENS, step=1, value=DEFAULT_MAX_NEW_TOKENS)
|
324 |
temperature = gr.Slider(label="Temperature", minimum=0.1, maximum=4.0, step=0.1, value=0.6)
|
325 |
top_p = gr.Slider(label="Top-p (nucleus sampling)", minimum=0.05, maximum=1.0, step=0.05, value=0.9)
|
326 |
top_k = gr.Slider(label="Top-k", minimum=1, maximum=1000, step=1, value=50)
|
327 |
repetition_penalty = gr.Slider(label="Repetition penalty", minimum=1.0, maximum=2.0, step=0.05, value=1.2)
|
328 |
-
|
329 |
with gr.Column():
|
330 |
-
output = gr.Textbox(label="Output", interactive=False, lines=
|
331 |
model_choice = gr.Radio(
|
332 |
-
choices=["Nanonets-OCR-s", "
|
333 |
label="Select Model",
|
334 |
value="Nanonets-OCR-s"
|
335 |
)
|
336 |
-
|
337 |
-
with gr.TabItem("Image Examples"):
|
338 |
-
gr.Examples(
|
339 |
-
examples=image_examples,
|
340 |
-
inputs=[image_query, image_upload, model_choice],
|
341 |
-
label="Click on an example to run"
|
342 |
-
)
|
343 |
-
with gr.TabItem("Video Examples"):
|
344 |
-
gr.Examples(
|
345 |
-
examples=video_examples,
|
346 |
-
inputs=[video_query, video_upload, model_choice],
|
347 |
-
label="Click on an example to run"
|
348 |
-
)
|
349 |
-
|
350 |
image_submit.click(
|
351 |
fn=generate_image,
|
352 |
inputs=[model_choice, image_query, image_upload, max_new_tokens, temperature, top_p, top_k, repetition_penalty],
|
@@ -359,4 +353,4 @@ with gr.Blocks(css=css, theme="bethecloud/storj_theme") as demo:
|
|
359 |
)
|
360 |
|
361 |
if __name__ == "__main__":
|
362 |
-
demo.queue(max_size=30).launch(share=True, ssr_mode=False, show_error=True)
|
|
|
285 |
|
286 |
# Define examples for image and video inference
|
287 |
image_examples = [
|
288 |
+
["Convert this page to docling", "images/1.png"],
|
289 |
+
["OCR the image", "images/2.jpg"],
|
290 |
+
["Convert this page to docling", "images/3.png"],
|
291 |
]
|
292 |
|
293 |
video_examples = [
|
|
|
307 |
|
308 |
# Create the Gradio Interface
|
309 |
with gr.Blocks(css=css, theme="bethecloud/storj_theme") as demo:
|
310 |
+
gr.Markdown("# **[OCR Net 4x](https://huggingface.co/collections/prithivMLmods/multimodal-implementations-67c9982ea04b39f0608badb0)**")
|
311 |
with gr.Row():
|
312 |
with gr.Column():
|
313 |
with gr.Tabs():
|
|
|
315 |
image_query = gr.Textbox(label="Query Input", placeholder="Enter your query here...")
|
316 |
image_upload = gr.Image(type="pil", label="Image")
|
317 |
image_submit = gr.Button("Submit", elem_classes="submit-btn")
|
318 |
+
gr.Examples(
|
319 |
+
examples=image_examples,
|
320 |
+
inputs=[image_query, image_upload, model_choice]
|
321 |
+
)
|
322 |
with gr.TabItem("Video Inference"):
|
323 |
video_query = gr.Textbox(label="Query Input", placeholder="Enter your query here...")
|
324 |
video_upload = gr.Video(label="Video")
|
325 |
video_submit = gr.Button("Submit", elem_classes="submit-btn")
|
326 |
+
gr.Examples(
|
327 |
+
examples=video_examples,
|
328 |
+
inputs=[video_query, video_upload]
|
329 |
+
)
|
330 |
with gr.Accordion("Advanced options", open=False):
|
331 |
max_new_tokens = gr.Slider(label="Max new tokens", minimum=1, maximum=MAX_MAX_NEW_TOKENS, step=1, value=DEFAULT_MAX_NEW_TOKENS)
|
332 |
temperature = gr.Slider(label="Temperature", minimum=0.1, maximum=4.0, step=0.1, value=0.6)
|
333 |
top_p = gr.Slider(label="Top-p (nucleus sampling)", minimum=0.05, maximum=1.0, step=0.05, value=0.9)
|
334 |
top_k = gr.Slider(label="Top-k", minimum=1, maximum=1000, step=1, value=50)
|
335 |
repetition_penalty = gr.Slider(label="Repetition penalty", minimum=1.0, maximum=2.0, step=0.05, value=1.2)
|
|
|
336 |
with gr.Column():
|
337 |
+
output = gr.Textbox(label="Output", interactive=False, lines=3, scale=2)
|
338 |
model_choice = gr.Radio(
|
339 |
+
choices=["Nanonets-OCR-s", "SmolDocling-256M-preview", "MonkeyOCR-Recognition"],
|
340 |
label="Select Model",
|
341 |
value="Nanonets-OCR-s"
|
342 |
)
|
343 |
+
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
344 |
image_submit.click(
|
345 |
fn=generate_image,
|
346 |
inputs=[model_choice, image_query, image_upload, max_new_tokens, temperature, top_p, top_k, repetition_penalty],
|
|
|
353 |
)
|
354 |
|
355 |
if __name__ == "__main__":
|
356 |
+
demo.queue(max_size=30).launch(share=True, mcp_server=True, ssr_mode=False, show_error=True)
|