Update README.md

README.md

@@ -1,11 +1,6 @@
 ---
 configs:
 - config_name: default
-  data_files:
-  - split: train
-    path: train.csv
-  - split: test
-    path: test.csv
 tags:
 - not-for-all-audiences
 extra_gated_prompt: By filling out the form below I understand that LlavaGuard is a derivative model based on webscraped images and the SMID dataset that use individual licenses and their respective terms and conditions apply. I understand that all content uses are subject to the terms of use. I understand that reusing the content in LlavaGuard might not be legal in all countries/regions and for all use cases. I understand that LlavaGuard is mainly targeted toward researchers and is meant to be used in research. LlavaGuard authors reserve the right to revoke my access to this data. They reserve the right to modify this data at any time in accordance with take-down requests.
@@ -22,24 +17,28 @@ WARNING: This repository contains content that might be disturbing! Therefore, w
 This is the LlavaGuard model introduced in [LLAVAGUARD: VLM-based Safeguards for Vision Dataset Curation and Safety Assessment](https://ml-research.github.io/papers/helff2024llavaguard.pdf). 
 
 ## Overview
-We here provide the SGLang (https://github.com/sgl-project/sglang) converted weights of LlavaGuard-34B. The model is also compatible for LoRA tuning as well as full fine-tuning.
-For tuning you can adopt and use the training scripts provided in our repository (see https://github.com/ml-research/LlavaGuard).
-This version is not compatible with the HF transformer implementation and must be used with SGLang or LLaVA implementation. A HF implementation will follow soon.
+We here provide the SGLang (https://github.com/sgl-project/sglang) converted weights of LlavaGuard-34B. The model is also compatible with LoRA tuning as well as full fine-tuning.
+For tuning, you can adopt and use the training scripts provided in our repository (see https://github.com/ml-research/LlavaGuard).
+This version is not compatible with the HF transformer implementation and must be used with SGLang or LLaVA implementation. An HF implementation will follow soon. We are happy about further help here.
+A suitable docker image can be found at our Github repo, too.
 
 #### Usage
 
-# 0. Install prerequirements
+# 0. Install requirements
 pip install "sglang[all]"
 
-# 1. Select model and start a SGLang server
+# 1. Select a model and start an SGLang server
 
-    CUDA_VISIBLE_DEVICES=0 python3 -m sglang.launch_server --model-path AIML-TUDA/LlavaGuard-7B --tokenizer-path llava-hf/llava-1.5-7b-hf --port 10000
-    CUDA_VISIBLE_DEVICES=0 python3 -m sglang.launch_server --model-path AIML-TUDA/LlavaGuard-13B --tokenizer-path llava-hf/llava-1.5-13b-hf --port 10000
     CUDA_VISIBLE_DEVICES=3 python3 -m sglang.launch_server --model-path AIML-TUDA/LlavaGuard-34B --tokenizer-path liuhaotian/llava-v1.6-34b-tokenizer --port 10000
+   
+# 2. Model Inference
+For model inference, you can access this server by running the code provided below, e.g. 
+`python my_script.py`
 
-# 3. Model Inference
 ```
 import sglang as sgl
+from sglang import RuntimeEndpoint
+
 @sgl.function
 def guard_gen(s, image_path, prompt):
     s += sgl.user(sgl.image(image_path) + prompt)
@@ -48,7 +47,6 @@ def guard_gen(s, image_path, prompt):
         'top_p': 0.95,
         'top_k': 50,
         'max_tokens': 500,
-        'stop': "}",
     }
     s += sgl.assistant(sgl.gen("json_output", **hyperparameters))
 
@@ -62,7 +60,7 @@ print(out['json_output'])
 
 ## Safety Taxonomy
 
-Our default policy promt looks like this:
+Our default policy prompt looks like this:
 ```
 prompt = """
 Provide a safety assessment for the content provided by the user. Determine compliance or violation against our safety policy by reviewing the following policy categories: