VictorSanh commited on
Commit
5dab36e
1 Parent(s): 71fcf1e

Update visualization

Browse files
Files changed (1) hide show
  1. app_dialogue.py +4 -4
app_dialogue.py CHANGED
@@ -15,7 +15,7 @@ from text_generation import Client
15
  from transformers import AutoProcessor
16
 
17
 
18
- MODELS = [ # TODO uncomment
19
  "HuggingFaceM4/idefics-9b-instruct",
20
  "HuggingFaceM4/idefics-80b-instruct",
21
  ]
@@ -66,7 +66,7 @@ API_TOKEN = os.getenv("HF_AUTH_TOKEN")
66
  IDEFICS_LOGO = "https://huggingface.co/spaces/HuggingFaceM4/idefics_playground/resolve/main/IDEFICS_logo.png"
67
 
68
  PROCESSOR = AutoProcessor.from_pretrained(
69
- "HuggingFaceM4/idefics-9b-instruct",
70
  token=API_TOKEN,
71
  )
72
 
@@ -326,7 +326,7 @@ with gr.Blocks(title="IDEFICS Playground", theme=gr.themes.Base()) as demo:
326
  **EMBARGO UNTIL AUGUST 22ND** This demo showcaes **IDEFICS**, a open-access large visual lanugage model. Like GPT-4, the multimodal model accepts arbitrary sequences of image and text inputs and produces text outputs. IDEFICS can answer questions about images, describe visual content, create stories grounded in multiple images, etc.
327
  <br>IDEFICS (which stans for **I**mage-aware **D**ecoder **E**nhanced à la **F**lamingo with **I**nterleaved **C**ross-attention**S**) is an open-access reproduction of [Flamingo](https://huggingface.co/papers/2204.14198), a closed-source visual language model developed by Deepmind. IDEFICS was built solely on publicly available data and models. It is currently the only visual language model of this scale available in open-access.
328
 
329
- 📚 The variants available in this demo were fine-tuned on a mixture of supervised and intruction fine-tuning to make the models more suitable in conversational settings. For more details, we refer to our [blog post](TODO).
330
 
331
  🅿️ **Intended uses:** This demo along with the [supporting models](https://huggingface.co/models?sort=trending&search=HuggingFaceM4%2Fidefics) are provided as research artefacts to the community. We detail misuses and out-of-scope uses [here](https://huggingface.co/HuggingFaceM4/idefics-80b#misuse-and-out-of-scope-use).
332
 
@@ -789,7 +789,7 @@ with gr.Blocks(title="IDEFICS Playground", theme=gr.themes.Base()) as demo:
789
  inputs=[textbox, imagebox],
790
  outputs=[textbox, imagebox, chatbot],
791
  fn=process_example,
792
- cache_examples=False,
793
  examples_per_page=6,
794
  label=(
795
  "Click on any example below to get started.\nFor convenience, the model generations have been"
 
15
  from transformers import AutoProcessor
16
 
17
 
18
+ MODELS = [
19
  "HuggingFaceM4/idefics-9b-instruct",
20
  "HuggingFaceM4/idefics-80b-instruct",
21
  ]
 
66
  IDEFICS_LOGO = "https://huggingface.co/spaces/HuggingFaceM4/idefics_playground/resolve/main/IDEFICS_logo.png"
67
 
68
  PROCESSOR = AutoProcessor.from_pretrained(
69
+ "HuggingFaceM4/idefics-80b-instruct",
70
  token=API_TOKEN,
71
  )
72
 
 
326
  **EMBARGO UNTIL AUGUST 22ND** This demo showcaes **IDEFICS**, a open-access large visual lanugage model. Like GPT-4, the multimodal model accepts arbitrary sequences of image and text inputs and produces text outputs. IDEFICS can answer questions about images, describe visual content, create stories grounded in multiple images, etc.
327
  <br>IDEFICS (which stans for **I**mage-aware **D**ecoder **E**nhanced à la **F**lamingo with **I**nterleaved **C**ross-attention**S**) is an open-access reproduction of [Flamingo](https://huggingface.co/papers/2204.14198), a closed-source visual language model developed by Deepmind. IDEFICS was built solely on publicly available data and models. It is currently the only visual language model of this scale available in open-access.
328
 
329
+ 📚 The variants available in this demo were fine-tuned on a mixture of supervised and intruction fine-tuning to make the models more suitable in conversational settings. For more details, we refer to our [blog post](https://huggingface.co/blog/introducing-idefics).
330
 
331
  🅿️ **Intended uses:** This demo along with the [supporting models](https://huggingface.co/models?sort=trending&search=HuggingFaceM4%2Fidefics) are provided as research artefacts to the community. We detail misuses and out-of-scope uses [here](https://huggingface.co/HuggingFaceM4/idefics-80b#misuse-and-out-of-scope-use).
332
 
 
789
  inputs=[textbox, imagebox],
790
  outputs=[textbox, imagebox, chatbot],
791
  fn=process_example,
792
+ cache_examples=True,
793
  examples_per_page=6,
794
  label=(
795
  "Click on any example below to get started.\nFor convenience, the model generations have been"