Updated README

README.md

@@ -1,51 +1,33 @@
 ---
 base_model: samrawal/problem-list-generator
-library_name: transformers
 tags:
 - llama-cpp
 - gguf-my-repo
+- medical
+- clinical
+language:
+- en
+pipeline_tag: text-generation
 ---
 
 # samrawal/problem-list-generator-Q8_0-GGUF
-This model was converted to GGUF format from [`samrawal/problem-list-generator`](https://huggingface.co/samrawal/problem-list-generator) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
-Refer to the [original model card](https://huggingface.co/samrawal/problem-list-generator) for more details on the model.
-
-## Use with llama.cpp
-Install llama.cpp through brew (works on Mac and Linux)
-
-```bash
-brew install llama.cpp
-
-```
-Invoke the llama.cpp server or the CLI.
-
-### CLI:
-```bash
-llama-cli --hf-repo samrawal/problem-list-generator-Q8_0-GGUF --hf-file problem-list-generator-q8_0.gguf -p "The meaning to life and the universe is"
-```
-
-### Server:
-```bash
-llama-server --hf-repo samrawal/problem-list-generator-Q8_0-GGUF --hf-file problem-list-generator-q8_0.gguf -c 2048
-```
-
-Note: You can also use this checkpoint directly through the [usage steps](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#usage) listed in the Llama.cpp repo as well.
-
-Step 1: Clone llama.cpp from GitHub.
-```
-git clone https://github.com/ggerganov/llama.cpp
-```
-
-Step 2: Move into the llama.cpp folder and build it with `LLAMA_CURL=1` flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).
-```
-cd llama.cpp && LLAMA_CURL=1 make
-```
-
-Step 3: Run inference through the main binary.
-```
-./llama-cli --hf-repo samrawal/problem-list-generator-Q8_0-GGUF --hf-file problem-list-generator-q8_0.gguf -p "The meaning to life and the universe is"
-```
-or 
-```
-./llama-server --hf-repo samrawal/problem-list-generator-Q8_0-GGUF --hf-file problem-list-generator-q8_0.gguf -c 2048
-```
+# About
+`problem-list-generator` is an experimental local LLM trained to generate a list of active clinical problems from a History and Physical (H&P) section of a clinical note. 
+
+![Problem List Generator Demo](https://github.com/samrawal/problem-list-generator/raw/main/assets/problemlistdemo.gif)
+
+
+This is part of [a small personal experiment](https://x.com/samarthrawal/status/1817686625364922582) to train small, fast, local language models that can assist clinicians when writing medical notes. It can be run completely offline, without any data leaving your local machine.
+
+For more information on trying this out, please see the [corresponding GitHub repository](https://github.com/samrawal/problem-list-generator).
+
+# Technical Details
+This model was finetuned using synthetic clinical data generated with `Llama 3.1 405b`. The model itself is a LORA finetune of `Meta-Llama-3.1-8B-Instruct`.
+
+This repository contains a basic Python script to inference the model via Ollama.
+
+# Setup
+1. Clone this repository
+1. Download the GGUF model from [here](https://huggingface.co/samrawal/problem-list-generator-Q8_0-GGUF) and place in same directory as the repo.
+3. Add the model to Ollama: `ollama create problemlist -f ProblemList.Modelfile`
+4. Run the Python script: `python problemlist.py`