prithivMLmods commited on
Commit
2e7d1b9
·
verified ·
1 Parent(s): a2404b9

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +8 -8
app.py CHANGED
@@ -96,10 +96,10 @@ def generate_image(model_name: str, text: str, image: Image.Image,
96
  """
97
  Generates responses using the selected model for image input.
98
  """
99
- if model_name == "VIREX-062225-exp":
100
  processor = processor_m
101
  model = model_m
102
- elif model_name == "DREX-062225-exp":
103
  processor = processor_x
104
  model = model_x
105
  elif model_name == "olmOCR-7B-0225":
@@ -152,10 +152,10 @@ def generate_video(model_name: str, text: str, video_path: str,
152
  """
153
  Generates responses using the selected model for video input.
154
  """
155
- if model_name == "VIREX-062225-exp":
156
  processor = processor_m
157
  model = model_m
158
- elif model_name == "DREX-062225-exp":
159
  processor = processor_x
160
  model = model_x
161
  elif model_name == "olmOCR-7B-0225":
@@ -275,17 +275,17 @@ with gr.Blocks(css=css, theme="bethecloud/storj_theme") as demo:
275
  markdown_output = gr.Markdown(label="Formatted Result (Result.Md)")
276
 
277
  model_choice = gr.Radio(
278
- choices=["DREX-062225-exp", "VIREX-062225-exp", "typhoon-ocr-3b", "olmOCR-7B-0225"],
279
  label="Select Model",
280
  value="DREX-062225-exp"
281
  )
282
 
283
  gr.Markdown("**Model Info 💻** | [Report Bug](https://huggingface.co/spaces/prithivMLmods/Doc-VLMs/discussions)")
284
- gr.Markdown("> [DREX-062225-exp](https://huggingface.co/prithivMLmods/DREX-062225-exp): the drex-062225-exp (document retrieval and extraction expert) model is a specialized fine-tuned version of docscopeocr-7b-050425-exp, optimized for document retrieval, content extraction, and analysis recognition. built on top of the qwen2.5-vl architecture.")
285
- gr.Markdown("> [VIREX-062225-exp](https://huggingface.co/prithivMLmods/VIREX-062225-exp): the virex-062225-exp (video information retrieval and extraction expert - experimental) model is a fine-tuned version of qwen2.5-vl-7b-instruct, specifically optimized for advanced video understanding, image comprehension, sense of reasoning, and natural language decision-making through cot reasoning.")
286
  gr.Markdown("> [typhoon-ocr-3b](https://huggingface.co/scb10x/typhoon-ocr-3b): a bilingual document parsing model built specifically for real-world documents in thai and english, inspired by models like olmocr, based on qwen2.5-vl-instruction. this model is intended to be used with a specific prompt only.")
287
  gr.Markdown("> [olmOCR-7B-0225](https://huggingface.co/allenai/olmOCR-7B-0225-preview): the olmocr-7b-0225-preview model is based on qwen2-vl-7b, optimized for document-level optical character recognition (ocr), long-context vision-language understanding, and accurate image-to-text conversion with mathematical latex formatting. designed with a focus on high-fidelity visual-textual comprehension.")
288
- gr.Markdown(">[!note] all the models in space are not guaranteed to perform well in video inference use cases.")
289
 
290
  image_submit.click(
291
  fn=generate_image,
 
96
  """
97
  Generates responses using the selected model for image input.
98
  """
99
+ if model_name == "VIREX-062225-8B-exp":
100
  processor = processor_m
101
  model = model_m
102
+ elif model_name == "DREX-062225-8B-exp":
103
  processor = processor_x
104
  model = model_x
105
  elif model_name == "olmOCR-7B-0225":
 
152
  """
153
  Generates responses using the selected model for video input.
154
  """
155
+ if model_name == "VIREX-062225-8B-exp":
156
  processor = processor_m
157
  model = model_m
158
+ elif model_name == "DREX-062225-8B-exp":
159
  processor = processor_x
160
  model = model_x
161
  elif model_name == "olmOCR-7B-0225":
 
275
  markdown_output = gr.Markdown(label="Formatted Result (Result.Md)")
276
 
277
  model_choice = gr.Radio(
278
+ choices=["DREX-062225-8B-exp", "VIREX-062225-8B-exp", "Typhoon-OCR-3B", "olmOCR-7B-0225"],
279
  label="Select Model",
280
  value="DREX-062225-exp"
281
  )
282
 
283
  gr.Markdown("**Model Info 💻** | [Report Bug](https://huggingface.co/spaces/prithivMLmods/Doc-VLMs/discussions)")
284
+ gr.Markdown("> [DREX-062225-8B-exp](https://huggingface.co/prithivMLmods/DREX-062225-exp): the drex-062225-exp (document retrieval and extraction expert) model is a specialized fine-tuned version of docscopeocr-7b-050425-exp, optimized for document retrieval, content extraction, and analysis recognition. built on top of the qwen2.5-vl architecture.")
285
+ gr.Markdown("> [VIREX-062225-8B-exp](https://huggingface.co/prithivMLmods/VIREX-062225-exp): the virex-062225-exp (video information retrieval and extraction expert - experimental) model is a fine-tuned version of qwen2.5-vl-7b-instruct, specifically optimized for advanced video understanding, image comprehension, sense of reasoning, and natural language decision-making through cot reasoning.")
286
  gr.Markdown("> [typhoon-ocr-3b](https://huggingface.co/scb10x/typhoon-ocr-3b): a bilingual document parsing model built specifically for real-world documents in thai and english, inspired by models like olmocr, based on qwen2.5-vl-instruction. this model is intended to be used with a specific prompt only.")
287
  gr.Markdown("> [olmOCR-7B-0225](https://huggingface.co/allenai/olmOCR-7B-0225-preview): the olmocr-7b-0225-preview model is based on qwen2-vl-7b, optimized for document-level optical character recognition (ocr), long-context vision-language understanding, and accurate image-to-text conversion with mathematical latex formatting. designed with a focus on high-fidelity visual-textual comprehension.")
288
+ gr.Markdown(">[⚠️note: all the models in space are not guaranteed to perform well in video inference use cases.")
289
 
290
  image_submit.click(
291
  fn=generate_image,