Text Generation
Transformers
Safetensors
English
cxrmate-2
chest X-ray report generation
radiology report generation
image captioning
chest X-ray
X-ray
radiology
cxrmate
cxrmate-ed
cxrmate-rrg24
report
radiology report
multimodal
patient data
mimic-cxr
custom_code
Instructions to use aehrc/cxrmate-2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use aehrc/cxrmate-2 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="aehrc/cxrmate-2", trust_remote_code=True)# Load model directly from transformers import AutoModelForCausalLM model = AutoModelForCausalLM.from_pretrained("aehrc/cxrmate-2", trust_remote_code=True, device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use aehrc/cxrmate-2 with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "aehrc/cxrmate-2" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "aehrc/cxrmate-2", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/aehrc/cxrmate-2
- SGLang
How to use aehrc/cxrmate-2 with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "aehrc/cxrmate-2" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "aehrc/cxrmate-2", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "aehrc/cxrmate-2" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "aehrc/cxrmate-2", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use aehrc/cxrmate-2 with Docker Model Runner:
docker model run hf.co/aehrc/cxrmate-2
Update dataset.py
Browse files- dataset.py +6 -3
dataset.py
CHANGED
|
@@ -12,8 +12,11 @@ class CXRMate2Dataset:
|
|
| 12 |
self.history = history
|
| 13 |
self.study_id_to_index = dict(zip(self.dataset['study_id'], range(len(self.dataset)), strict=True))
|
| 14 |
|
| 15 |
-
def __getitem__(self,
|
| 16 |
-
|
|
|
|
|
|
|
|
|
|
| 17 |
|
| 18 |
if 'views' not in batch:
|
| 19 |
batch['views'] = [None] * len(batch['images'])
|
|
@@ -26,7 +29,7 @@ class CXRMate2Dataset:
|
|
| 26 |
|
| 27 |
if self.history:
|
| 28 |
|
| 29 |
-
if
|
| 30 |
|
| 31 |
# Sort by datetime to ensure correct order:
|
| 32 |
assert all(i is not None and not (isinstance(i, float) and np.isnan(i)) for i in batch['prior_study_datetimes'])
|
|
|
|
| 12 |
self.history = history
|
| 13 |
self.study_id_to_index = dict(zip(self.dataset['study_id'], range(len(self.dataset)), strict=True))
|
| 14 |
|
| 15 |
+
def __getitem__(self, key):
|
| 16 |
+
if not isinstance(key, int):
|
| 17 |
+
return self.dataset[key]
|
| 18 |
+
|
| 19 |
+
batch = self.dataset[key]
|
| 20 |
|
| 21 |
if 'views' not in batch:
|
| 22 |
batch['views'] = [None] * len(batch['images'])
|
|
|
|
| 29 |
|
| 30 |
if self.history:
|
| 31 |
|
| 32 |
+
if batch['prior_study_ids']:
|
| 33 |
|
| 34 |
# Sort by datetime to ensure correct order:
|
| 35 |
assert all(i is not None and not (isinstance(i, float) and np.isnan(i)) for i in batch['prior_study_datetimes'])
|