Update README.md

README.md

@@ -1,199 +1,436 @@
----
+```
 library_name: transformers
-tags: []
----
-
-# Model Card for Model ID
-
-<!-- Provide a quick summary of what the model is/does. -->
-
-
+tags:
+- medical
+- q&a
+- pubmedqa
+- diffusiongemma
+- lora
+- unsloth
+license: apache-2.0
+datasets:
+- qiaojin/PubMedQA
+language:
+- en
+metrics:
+- accuracy
+base_model:
+- google/diffusiongemma-26B-A4B-it
+pipeline_tag: text-generation
+```
+
+# DiffusionGemma PubMedQA
+
+This model is a LoRA fine-tuned version of **DiffusionGemma 26B-A4B IT** for biomedical question answering on **PubMedQA**.
+
+The model was fine-tuned to answer biomedical research questions using PubMed abstract context and return one of three labels:
+
+```text
+yes / no / maybe
+```
+
+> This model is intended for experimentation, benchmarking, and educational use. It is **not** intended for clinical decision-making or medical advice.
 
 ## Model Details
 
 ### Model Description
 
-<!-- Provide a longer summary of what this model is. -->
+This model adapts DiffusionGemma to the PubMedQA task. Given a biomedical research question and supporting PubMed abstract context, the model predicts whether the answer is `yes`, `no`, or `maybe`.
 
-This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
+- **Developed by:** kingabzpro
+- **Shared by:** kingabzpro
+- **Model type:** Diffusion language model with LoRA adapter
+- **Language(s):** English
+- **License:** Apache 2.0
+- **Fine-tuned from:** `google/diffusiongemma-26B-A4B-it`
+- **Training framework:** Unsloth + Transformers
+- **Task:** Biomedical question answering / text generation
+- **Dataset:** `qiaojin/PubMedQA`
 
-- **Developed by:** [More Information Needed]
-- **Funded by [optional]:** [More Information Needed]
-- **Shared by [optional]:** [More Information Needed]
-- **Model type:** [More Information Needed]
-- **Language(s) (NLP):** [More Information Needed]
-- **License:** [More Information Needed]
-- **Finetuned from model [optional]:** [More Information Needed]
+### Model Sources
 
-### Model Sources [optional]
+- **Repository:** `kingabzpro/diffusiongemma_pubmedqa`
+- **Base model:** `google/diffusiongemma-26B-A4B-it`
+- **Dataset:** `qiaojin/PubMedQA`
 
-<!-- Provide the basic links for the model. -->
+## Uses
 
-- **Repository:** [More Information Needed]
-- **Paper [optional]:** [More Information Needed]
-- **Demo [optional]:** [More Information Needed]
+### Direct Use
 
-## Uses
+This model can be used to answer PubMedQA-style biomedical research questions where the input includes:
 
-<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+1. A biomedical research question
+2. Relevant abstract/context text
+3. A request to answer with `yes`, `no`, or `maybe`
 
-### Direct Use
+Example task format:
+
+```text
+Answer the biomedical research question using only the context.
+
+Context:
+[PubMed abstract context]
 
-<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+Question:
+[Biomedical research question]
 
-[More Information Needed]
+Answer with only one word: yes, no, or maybe.
+```
 
-### Downstream Use [optional]
+### Downstream Use
 
-<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+This model may be useful for:
 
-[More Information Needed]
+- Biomedical QA experiments
+- PubMedQA-style benchmark testing
+- Fine-tuning tutorials
+- LoRA adapter experiments with DiffusionGemma
+- Educational demos for medical-domain model adaptation
 
 ### Out-of-Scope Use
 
-<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+This model should **not** be used for:
 
-[More Information Needed]
+- Medical diagnosis
+- Treatment recommendations
+- Emergency medical advice
+- Replacing a doctor, pharmacist, or clinical expert
+- Patient-specific medical decisions
+- High-stakes biomedical or healthcare deployment without further validation
+
+The model is trained on a narrow benchmark-style task and may produce incorrect answers.
 
 ## Bias, Risks, and Limitations
 
-<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+This model has several important limitations:
 
-[More Information Needed]
+- It was fine-tuned on PubMedQA-style examples, not general medical conversations.
+- The model predicts only `yes`, `no`, or `maybe`, so it may oversimplify complex biomedical findings.
+- The evaluation set used in this experiment was small: 50 examples.
+- The model may be sensitive to prompt format.
+- The model may answer incorrectly if the context is incomplete, misleading, or unrelated.
+- The training target was short, so the loss dropped quickly and may not reflect deep medical reasoning.
+- The model should not be treated as medically reliable.
 
 ### Recommendations
 
-<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+Users should:
 
-Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+- Use this model only for research and educational experiments.
+- Always verify outputs against trusted biomedical sources.
+- Avoid using the model for real clinical or patient-facing decisions.
+- Run larger evaluations before drawing strong conclusions.
+- Consider training on explanations, not only one-word labels, for a more meaningful medical QA setup.
 
 ## How to Get Started with the Model
 
-Use the code below to get started with the model.
+### Install dependencies
+
+```bash
+pip install unsloth
+pip install transformers datasets peft accelerate sentencepiece protobuf
+```
+
+### Load the model
+
+If this repository contains the LoRA adapter, load the base model first and then attach the adapter:
+
+```python
+import copy
+import torch
+from peft import PeftModel
+from unsloth import FastModel
+
+base_model_name = "unsloth/diffusiongemma-26B-A4B-it"
+adapter_name = "kingabzpro/diffusiongemma_pubmedqa"
+
+model, tokenizer = FastModel.from_pretrained(
+    model_name=base_model_name,
+    dtype=torch.bfloat16,
+    load_in_4bit=False,
+)
+
+model = PeftModel.from_pretrained(model, adapter_name)
+
+processor = tokenizer
+tok = processor.tokenizer if hasattr(processor, "tokenizer") else processor
+
+dev = next(
+    (p.device for p in model.parameters() if p.device.type != "meta"),
+    torch.device("cuda"),
+)
+
+canvas_len = model.config.canvas_length
+```
+
+### Run inference
+
+```python
+def answer_question(prompt, steps=16):
+    input_ids = processor.apply_chat_template(
+        [{"role": "user", "content": prompt}],
+        tokenize=True,
+        add_generation_prompt=True,
+        return_tensors="pt",
+    ).to(dev)
+
+    gen_config = copy.deepcopy(model.generation_config)
+    gen_config.max_denoising_steps = steps
+    gen_config.max_new_tokens = canvas_len
+
+    model.eval()
+
+    with torch.no_grad():
+        output = model.generate(
+            input_ids=input_ids,
+            generation_config=gen_config,
+        )
 
-[More Information Needed]
+    generated = output.sequences[0, input_ids.shape[1]:]
+    text = tok.decode(generated.tolist(), skip_special_tokens=True)
+
+    return text.strip().lower()
+
+
+prompt = """Answer the biomedical research question using only the context.
+
+Context:
+[Paste PubMed abstract context here]
+
+Question:
+[Paste biomedical question here]
+
+Answer with only one word: yes, no, or maybe."""
+
+print(answer_question(prompt, steps=16))
+```
 
 ## Training Details
 
 ### Training Data
 
-<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+The model was fine-tuned on `qiaojin/PubMedQA`.
 
-[More Information Needed]
+The notebook used:
+
+- **Training split:** `pqa_artificial`
+- **Evaluation split:** `pqa_labeled`
+- **Training examples used:** 3,000
+- **Evaluation examples prepared:** 200
+- **Evaluation examples used for reported result:** 50
+
+Each training example was converted into a prompt-answer pair:
+
+```text
+Input:
+Biomedical question + PubMed abstract context
+
+Target:
+yes / no / maybe
+```
 
 ### Training Procedure
 
-<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+#### Preprocessing
 
-#### Preprocessing [optional]
+For each PubMedQA row:
 
-[More Information Needed]
+1. The abstract contexts were joined into one context block.
+2. The context was truncated to 2,500 characters.
+3. The question was inserted below the context.
+4. The target answer was the `final_decision` field.
+5. Only examples with `yes`, `no`, or `maybe` labels were used.
 
+Prompt format:
 
-#### Training Hyperparameters
+```text
+Answer the biomedical research question using only the context.
 
-- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+Context:
+{context}
 
-#### Speeds, Sizes, Times [optional]
+Question:
+{question}
 
-<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+Answer with only one word: yes, no, or maybe.
+```
 
-[More Information Needed]
+Target format:
 
-## Evaluation
+```text
+{final_decision}
+```
 
-<!-- This section describes the evaluation protocols and provides the results. -->
+#### Training Hyperparameters
+
+- **Training regime:** bf16
+- **LoRA rank:** 64
+- **LoRA alpha:** 128
+- **Trainable parameters:** 149,630,976
+- **Total parameters:** 25,973,409,840
+- **Trainable percentage:** 0.5761%
+- **Training examples:** 3,000
+- **Training steps:** 60
+- **Gradient accumulation:** 4
+- **Learning rate:** 1e-4
+- **Optimizer:** AdamW
+- **Scheduler:** OneCycleLR
+- **Weight decay:** 0.0
+- **Max context characters:** 2,500
+- **Canvas length:** 256
+- **Dataset:** `qiaojin/PubMedQA`
+
+#### Speeds, Sizes, Times
+
+Training was run on a RunPod H100 notebook.
+
+Training logs from the saved notebook:
+
+```text
+step 20/60 | loss 0.0019 | 43s
+step 40/60 | loss 0.0003 | 85s
+step 60/60 | loss 0.0001 | 126s
+```
+
+Approximate training time:
+
+```text
+126 seconds for 60 steps
+```
+
+## Evaluation
 
 ### Testing Data, Factors & Metrics
 
 #### Testing Data
 
-<!-- This should link to a Dataset Card if possible. -->
+Evaluation used the `pqa_labeled` subset of `qiaojin/PubMedQA`.
 
-[More Information Needed]
+The reported run used:
 
-#### Factors
+- **Evaluation examples:** 50
+- **Denoising steps:** 16
+- **Metric:** Accuracy
 
-<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+#### Factors
 
-[More Information Needed]
+The evaluation was not disaggregated by biomedical topic, article type, answer class, or question type. Results should be treated as a small sanity-check evaluation, not a full benchmark.
 
 #### Metrics
 
-<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+Accuracy was used because PubMedQA final decisions are discrete labels:
 
-[More Information Needed]
+```text
+yes / no / maybe
+```
+
+A prediction was counted as correct if the cleaned model output matched the gold `final_decision`.
 
 ### Results
 
-[More Information Needed]
+| Setting | Accuracy | Correct / Total |
+| --- | --- | --- |
+| Before fine-tuning | 0.60 | 30 / 50 |
+| After fine-tuning | 0.80 | 40 / 50 |
+| Improvement | +0.20 | +10 / 50 |
 
 #### Summary
 
+In the saved RunPod H100 notebook run, the model improved from **60% accuracy before fine-tuning** to **80% accuracy after fine-tuning** on a 50-example PubMedQA evaluation sample.
 
+This is a **+20 percentage point improvement**.
 
-## Model Examination [optional]
+The result shows that the model can quickly adapt to the PubMedQA answer format. However, this is a small evaluation and should not be interpreted as a clinically meaningful benchmark.
 
-<!-- Relevant interpretability work for the model goes here -->
+## Model Examination
 
-[More Information Needed]
+No detailed interpretability or model examination was performed.
 
 ## Environmental Impact
 
-<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
-
-Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+Carbon emissions were not measured for this run.
 
-- **Hardware Type:** [More Information Needed]
-- **Hours used:** [More Information Needed]
-- **Cloud Provider:** [More Information Needed]
-- **Compute Region:** [More Information Needed]
-- **Carbon Emitted:** [More Information Needed]
+- **Hardware Type:** NVIDIA H100 80GB HBM3
+- **Hours used:** Approximately 0.04 hours for the 60-step training loop, excluding setup, model loading, and evaluation
+- **Cloud Provider:** RunPod
+- **Compute Region:** Not recorded
+- **Carbon Emitted:** Not measured
 
-## Technical Specifications [optional]
+## Technical Specifications
 
 ### Model Architecture and Objective
 
-[More Information Needed]
+The base model is DiffusionGemma 26B-A4B IT, a diffusion-style language model. The fine-tuning used LoRA adapters.
 
-### Compute Infrastructure
+The training objective followed a block-diffusion setup:
 
-[More Information Needed]
+1. Encode the target answer into the model canvas.
+2. Randomly corrupt answer tokens.
+3. Train the model to reconstruct the clean answer.
+4. Apply loss only over the target answer tokens.
+
+### Compute Infrastructure
 
 #### Hardware
 
-[More Information Needed]
+- NVIDIA H100 80GB HBM3
+- Reported GPU memory: approximately 85 GB total
 
 #### Software
 
-[More Information Needed]
+- Python
+- PyTorch 2.10.0+cu128
+- Transformers
+- Unsloth
+- Unsloth Zoo
+- PEFT
+- Datasets
+- RunPod Jupyter Notebook
+
+## Citation
 
-## Citation [optional]
+If you use this model, please cite the original PubMedQA dataset and DiffusionGemma base model.
 
-<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+**PubMedQA:**
 
-**BibTeX:**
+```bibtex
+@inproceedings{jin2019pubmedqa,
+  title={PubMedQA: A Dataset for Biomedical Research Question Answering},
+  author={Jin, Qiao and Dhingra, Bhuwan and Liu, Zhengping and Cohen, William W. and Lu, Xinghua},
+  booktitle={Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing},
+  year={2019}
+}
+```
 
-[More Information Needed]
+## Glossary
 
-**APA:**
+- **PubMedQA:** A biomedical question-answering dataset based on PubMed abstracts.
+- **LoRA:** Low-Rank Adaptation, a parameter-efficient fine-tuning method.
+- **DiffusionGemma:** A diffusion-style language model.
+- **Denoising steps:** Iterative generation steps used by diffusion models.
+- **Accuracy:** Percentage of predictions matching the gold label.
 
-[More Information Needed]
+## More Information
 
-## Glossary [optional]
+This model was created as a simple fine-tuning experiment for adapting DiffusionGemma to a medical QA dataset.
 
-<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+The task is intentionally simple:
 
-[More Information Needed]
+```text
+Biomedical context + question → yes / no / maybe
+```
 
-## More Information [optional]
+For a stronger medical QA model, future versions should train on both:
 
-[More Information Needed]
+```text
+Decision: yes/no/maybe
+Explanation: short evidence-based explanation
+```
 
-## Model Card Authors [optional]
+## Model Card Authors
 
-[More Information Needed]
+- kingabzpro
 
 ## Model Card Contact
 
-[More Information Needed]
+For questions, contact the model repository owner on Hugging Face.