Update README.md
Browse files
README.md
CHANGED
|
@@ -46,10 +46,6 @@ The model is specifically fine-tuned to automate cardiological recommendations a
|
|
| 46 |
### 1. Requirements
|
| 47 |
To run this model, you need [llama.cpp](https://github.com/ggerganov/llama.cpp) or any GGUF-compatible software (LM Studio, Jan, Ollama).
|
| 48 |
|
| 49 |
-
## 🚀 Try the model online (ukrainian)
|
| 50 |
-
You can test the model's performance directly in your browser:
|
| 51 |
-
[**CardioGemma Demo**](https://test-amosov.esemi.org/) (External) or via [Hugging Face Spaces](https://huggingface.co/spaces/uaritm/cardiogemma).
|
| 52 |
-
|
| 53 |
### 2. Recommended Inference Settings
|
| 54 |
Based on the fine-tuning process, the following parameters are recommended for medical accuracy:
|
| 55 |
- **Temperature:** 0.01 (near-deterministic)
|
|
|
|
| 46 |
### 1. Requirements
|
| 47 |
To run this model, you need [llama.cpp](https://github.com/ggerganov/llama.cpp) or any GGUF-compatible software (LM Studio, Jan, Ollama).
|
| 48 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 49 |
### 2. Recommended Inference Settings
|
| 50 |
Based on the fine-tuning process, the following parameters are recommended for medical accuracy:
|
| 51 |
- **Temperature:** 0.01 (near-deterministic)
|