Upload folder using huggingface_hub

README.md

@@ -1,12 +1,54 @@
-# K2-Inhale (LoRA Adapter)
-
-**Base model:** `LLM360/K2-Think`  
-**Author fine-tune:** `SutanRifkyt`  
-**Method:** QLoRA (4-bit)  
-**Domain:** Lung CT & PET/CT findings (nodule, consolidation, FDG uptake, staging hints)  
-**Use case:** Explain findings in plain language for patients, say how concerning for lung cancer, and suggest next step (follow-up CT, PET-CT, biopsy, urgent oncologist etc).
-
-## How to use
+---
+library_name: peft
+base_model: LLM360/K2-Think
+pipeline_tag: text-generation
+license: apache-2.0
+datasets:
+  - TCIA
+  - internal_synthetic_clinical_like_reports
+language:
+  - en
+  - id
+tags:
+  - medical
+  - lung-ct
+  - pet-ct
+  - oncology
+  - triage
+  - peft
+  - qlora
+model_name: K2-Inhale (QLoRA adapter)
+inference: false
+---
+
+# K2-Inhale 🫁 (LoRA Adapter for LLM360/K2-Think)
+
+**Author / Fine-tune:** Sutan Rifky Tedjasukmana (@SutanRifkyt)  
+**Base model:** `LLM360/K2-Think` (credit to LLM360)  
+**Method:** QLoRA (4-bit base) with PEFT adapters  
+**Domain:** Lung CT & PET/CT findings — nodules, consolidation, FDG uptake, possible staging hints  
+**Languages:** English + Bahasa Indonesia (output is patient-friendly, non-radiologist tone)  
+**Intended use:** Patient-facing explanation + triage suggestion  
+**Not intended for:** Final diagnosis, treatment planning, or replacing licensed clinicians.
+
+---
+
+## 🔍 What this model does
+
+K2-Inhale is a lightweight LoRA adapter trained on top of `LLM360/K2-Think` to:
+1. Rewrite lung CT / PET-CT findings into patient-friendly explanation.
+2. Give a plain-language "how worrying is this for lung cancer".
+3. Suggest a next step (follow-up CT, PET-CT, tissue biopsy, urgent oncologist, etc.).
+
+Target audience is:
+- patients who just got an imaging report and are anxious,
+- junior clinicians who want a patient-facing summary first draft.
+
+⚠️ This model is **NOT** a medical device and should **NOT** be used for autonomous diagnosis.
+
+---
+
+## 🧠 How to load (recommended path = base model + LoRA)
 
 ```python
 import torch
@@ -29,7 +71,12 @@ model = AutoModelForCausalLM.from_pretrained(
     trust_remote_code=False
 )
 
-model = PeftModel.from_pretrained(model, adapter_id)
+model = PeftModel.from_pretrained(
+    model,
+    adapter_id,
+    torch_dtype=torch.bfloat16,
+    device_map="auto",
+)
 
 prompt = """<|user|>:
 Explain this chest CT finding in simple language for the patient, assess how concerning it is for lung cancer, and say what should happen next.
@@ -45,33 +92,74 @@ with torch.no_grad():
     output = model.generate(
         **inputs,
         max_new_tokens=300,
-        do_sample=False,
-        temperature=0.0,
+        temperature=0.2,
+        do_sample=True,
     )
 
 print(tokenizer.decode(output[0], skip_special_tokens=True))
-Training data
+⚡ Quantized version
+
+For easier inference on smaller GPUs / single consumer cards, a quantized export is included under quantized/.
+
+quantized/ is an experimental merged model snapshot intended for local testing / demo.
+Quality may be lower vs full base+LoRA above.
+
+Basic usage (example, adjust to your runtime):
+
+import torch
+from transformers import AutoTokenizer, AutoModelForCausalLM
+
+quantized_id = "SutanRifkyt/K2-Inhale/quantized"
+
+tokenizer = AutoTokenizer.from_pretrained(
+    quantized_id,
+    use_fast=False,
+    trust_remote_code=False
+)
+
+model = AutoModelForCausalLM.from_pretrained(
+    quantized_id,
+    torch_dtype=torch.float16,
+    device_map="auto",
+    trust_remote_code=False
+)
+
+
+Note: If you see GGUF / AWQ / bitsandbytes entries, load with the correct loader for that format.
+
+📚 Training data (high-level)
+
+~8k supervised instruction-style pairs constructed from:
+
+public lung CT and PET/CT descriptions (incl. TCIA-like oncology cohorts),
+
+synthetic expansions of impression/assessment text,
+
+staged "what happens next" counseling scripts.
+
+Each sample looks like:
+
+instruction: "Explain this finding for the patient, include cancer concern level, and next step"
 
-Merged ~8k instruction-style samples from:
+input: actual CT/PET-CT style text (nodule size, FDG uptake, etc.)
 
-ReXGroundingCT style CT findings (free-text localized abnormalities)
+output: step-by-step reasoning and final recommendation in plain language.
 
-Lung-PET-CT-Dx (TCIA) PET/CT cases with histopathology labels and staging clues
+🚨 Safety & limitations
 
-Each sample is turned into:
+This model is for triage / education, not diagnosis.
 
-instruction: ask model to explain for patient, assess cancer concern, propose next step
+It may sound confident even when uncertain.
 
-input: actual radiology-style finding text
+It has not been clinically validated.
 
-output: chain-of-thought style reasoning + final recommendation
+Always involve a radiologist / oncologist for real decisions.
 
-Safety
+✍️ Citation / credit
 
-This model is not a doctor. It's a triage / education assistant.
-It should not replace radiologist, oncologist, or clinical decision-making.
+Base model LLM360/K2-Think is released by the LLM360 team.
+This repository only publishes LoRA/PEFT adapter weights and an optional quantized snapshot, fine-tuned by Sutan Rifky Tedjasukmana (@SutanRifkyt) for lung imaging triage.
 
-License
+License: Apache-2.0 for adapter weights.
+Underlying medical text sources may include portions of CC BY 4.0 datasets and synthetic expansions derived from them.
 
-LoRA weights are provided for research use.
-Source data includes CC BY 4.0 material from The Cancer Imaging Archive (TCIA) and academic datasets.