Update README.md

README.md

@@ -1,199 +1,261 @@
 ---
+language: en
 library_name: transformers
-tags: []
+pipeline_tag: text-classification
+tags:
+  - text-classification
+  - sequence-classification
+  - roberta
+  - distilroberta
+  - climate-change
+  - logical-fallacy-detection
+  - nlp
+license: apache-2.0
+model-index:
+  - name: climate-fallacy-roberta
+    results:
+      - task:
+          type: text-classification
+          name: Climate logical fallacy classification
+        dataset:
+          name: Climate subset of Tariq60/fallacy-detection
+          type: custom
+          split: test
+        metrics:
+          - name: Accuracy
+            type: accuracy
+            value: 0.24
+          - name: Macro F1
+            type: f1
+            value: 0.20
+          - name: Weighted F1
+            type: f1
+            value: 0.24
 ---
 
-# Model Card for Model ID
-
-<!-- Provide a quick summary of what the model is/does. -->
+# Climate Logical Fallacy Classifier (DistilRoBERTa)
 
+This model is a **DistilRoBERTa**–based text classification model fine-tuned to detect **logical fallacies in climate-related text**.  
+It predicts one of 11 logical fallacy labels (including “NO_FALLACY”) for a given sentence or short paragraph.
 
+The model was trained as part of an academic NLP project on _“Automated Detection of Logical Fallacies in Climate Change Social Media Posts using Small Language Models (SLMs)”_.
 
 ## Model Details
 
-### Model Description
-
-<!-- Provide a longer summary of what this model is. -->
-
-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:** [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 [optional]
-
-<!-- Provide the basic links for the model. -->
-
-- **Repository:** [More Information Needed]
-- **Paper [optional]:** [More Information Needed]
-- **Demo [optional]:** [More Information Needed]
-
-## Uses
-
-<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
-
-### Direct Use
+- **Base model**: `distilroberta-base`
+- **Architecture**: DistilRoBERTa (Transformer encoder, 6 layers)
+- **Task**: Multi-class text classification
+- **Number of classes**: 11
+- **Language**: English
+- **Framework**:  Transformers
 
-<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+### Label Set
 
-[More Information Needed]
+The model is trained to predict the following labels:
 
-### Downstream Use [optional]
+1. `CHERRY_PICKING`
+2. `EVADING_THE_BURDEN_OF_PROOF`
+3. `FALSE_ANALOGY`
+4. `FALSE_AUTHORITY`
+5. `FALSE_CAUSE`
+6. `HASTY_GENERALISATION`
+7. `NO_FALLACY`
+8. `POST_HOC`
+9. `RED_HERRINGS`
+10. `STRAWMAN`
+11. `VAGUENESS`
 
-<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+`id2label` / `label2id` mappings are stored in the model config and are consistent with the training code.
 
-[More Information Needed]
+## 📚 Training Data
 
-### Out-of-Scope Use
+The model was fine-tuned on the **climate subset** of the open-source dataset from:
 
-<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+> Tariq60 – *fallacy-detection* repository  
+> https://github.com/Tariq60/fallacy-detection
 
-[More Information Needed]
+Only the **climate** portion of the dataset was used, with the standard split:
 
-## Bias, Risks, and Limitations
+- `train/` – training examples
+- `dev/` – validation examples
+- `test/` – held-out evaluation set
 
-<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+Each example includes:
 
-[More Information Needed]
+- The climate-related text segment
+- A manually assigned fallacy label (or `No fallacy`)
 
-### Recommendations
+### Preprocessing
 
-<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+- Texts were lower-cased and cleaned using a light `basic_clean` function:
+  - Stripping extra whitespace
+  - Normalising some punctuation
+- Some classes were **minority labels** (few examples), so basic **class balancing** was applied via up-sampling in the training set.
+- NaN or empty texts were dropped before training.
 
-Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+## Training Procedure
 
-## How to Get Started with the Model
+- **Base model**: `distilroberta-base`
+- **Optimizer**: AdamW (via `Trainer`)
+- **Learning rate**: 2e-5
+- **Batch size**: 16
+- **Max sequence length**: 128–256 tokens
+- **Epochs**: 10
+- **Weight decay**: 0.01
+- **Loss function**: Cross-entropy, optionally with class weights to mitigate class imbalance
+- **Validation split**: 80/20 stratified split of the training data
 
-Use the code below to get started with the model.
+## Implementation used:
 
-[More Information Needed]
+- `AutoTokenizer`
+- `AutoModelForSequenceClassification`
+- `TrainingArguments`
+- `Trainer`
 
-## 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. -->
-
-[More Information Needed]
-
-### 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 [optional]
-
-[More Information Needed]
+from the Transformers library.
 
+## Evaluation
 
-#### Training Hyperparameters
+Evaluation was done on the **held-out climate test set** from the dataset.
 
-- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+**Metrics (multi-class):**
 
-#### Speeds, Sizes, Times [optional]
+- **Accuracy** ≈ 0.24  
+- **Macro F1** ≈ 0.20  
+- **Weighted F1** ≈ 0.24  
 
-<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+These values are **baseline experimental results** on a relatively small and imbalanced dataset. They should be interpreted as *preliminary research numbers*, not as production-ready performance.
 
-[More Information Needed]
+Different random seeds, data balancing strategies, or more aggressive hyperparameter tuning can change these numbers.
 
-## Evaluation
+## Intended Use
 
-<!-- This section describes the evaluation protocols and provides the results. -->
+### Primary Use
 
-### Testing Data, Factors & Metrics
+- Research and experimentation on:
+  - Automated detection of logical fallacies in climate discourse
+  - Comparing traditional baselines (TF-IDF + SVM) vs. Transformer-based models
+  - Building educational tools that flag potential fallacies in climate arguments
 
-#### Testing Data
+### Suitable Scenarios
 
-<!-- This should link to a Dataset Card if possible. -->
+- Analyzing **short climate-related social media posts**  
+- Demonstration / teaching examples on:
+  - Argumentation quality
+  - Climate misinformation
+  - Explainable NLP (combined with a small language model explainer, e.g. FLAN-T5)
+  - 
+## Limitations & Ethical Considerations
 
-[More Information Needed]
+### Limitations
 
-#### Factors
+- **Small dataset**: Training data is limited in size, especially for rarer fallacy types.
+- **Class imbalance**: Some fallacies occur far less frequently, which affects per-class F1 scores.
+- **Modest performance**: Overall accuracy and macro F1 are relatively low. The model should be treated as an exploratory research artifact, not a production system.
+- **Domain specificity**: The model is trained only on **climate** discourse; performance on other topics (e.g. politics, health) is unknown and likely poor.
 
-<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+### Ethical Considerations
 
-[More Information Needed]
+- Predictions are **probabilistic**, not definitive judgments of truth or deception.
+- The model can be **wrong or over-confident**, especially on borderline or nuanced arguments.
+- It should **not** be used for automated moderation, censorship, or any high-stakes decision-making without strong human oversight.
 
-#### Metrics
+##  How to Integration with Explanatory SLM
 
-<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+In the associated project, this classifier is combined with a small language model (e.g., google/flan-t5-small) to generate natural-language explanations of the predicted fallacy label:
 
-[More Information Needed]
+What the fallacy means in simple terms
 
-### Results
+Why the input text might be an example
 
-[More Information Needed]
+This setup is used in a Streamlit app:
 
-#### Summary
+Users enter a climate-related argument
 
+The model predicts a fallacy label
 
+FLAN-T5 generates a short explanation
 
-## Model Examination [optional]
+## Citation
 
-<!-- Relevant interpretability work for the model goes here -->
+If you use this model in academic work, you can cite it as:
 
-[More Information Needed]
+Kyeremeh, F. (2025). Climate Logical Fallacy Classifier (DistilRoBERTa). Hugging Face.
+Model: SteadyHands/climate-fallacy-roberta.
 
-## Environmental Impact
+And also consider citing the original dataset author(s):
 
-<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+Tariq60. fallacy-detection GitHub repository.
+https://github.com/Tariq60/fallacy-detection
 
-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).
+## Acknowledgements
 
-- **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]
+Base model: distilroberta-base by Hugging Face
 
-## Technical Specifications [optional]
+Dataset: Climate subset from Tariq60’s fallacy-detection repository
 
-### Model Architecture and Objective
+## Libraries:
 
-[More Information Needed]
+Transformers
 
-### Compute Infrastructure
+ Datasets
 
-[More Information Needed]
+scikit-learn
 
-#### Hardware
+## Project context: 
 
-[More Information Needed]
+Master ’s-level NLP / Data Science coursework on Small Language Models and explainable NLP.
 
-#### Software
+##  How to Use
 
-[More Information Needed]
+### Python Example (Logits → Label)
 
-## Citation [optional]
+```python
+from transformers import AutoTokenizer, AutoModelForSequenceClassification
+import torch
 
-<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+model_id = "SteadyHands/climate-fallacy-roberta"
 
-**BibTeX:**
+tokenizer = AutoTokenizer.from_pretrained(model_id)
+model = AutoModelForSequenceClassification.from_pretrained(model_id)
 
-[More Information Needed]
+text = "Climate has always changed in the past, so current warming can't be caused by humans."
 
-**APA:**
+inputs = tokenizer(
+    text,
+    return_tensors="pt",
+    truncation=True,
+    padding="max_length",
+    max_length=256,
+)
 
-[More Information Needed]
+with torch.no_grad():
+    outputs = model(**inputs)
 
-## Glossary [optional]
+logits = outputs.logits
+probs = torch.softmax(logits, dim=-1)[0].tolist()
+pred_id = int(torch.argmax(logits, dim=-1).item())
 
-<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+id2label = model.config.id2label
+pred_label = id2label[str(pred_id)] if isinstance(id2label, dict) else id2label[pred_id]
 
-[More Information Needed]
+print("Text:", text)
+print("Predicted label:", pred_label)
+print("Probabilities:", probs)
 
-## More Information [optional]
 
-[More Information Needed]
+Using the Transformers Pipeline
 
-## Model Card Authors [optional]
+```python
+from transformers import pipeline
 
-[More Information Needed]
+clf = pipeline(
+    "text-classification",
+    model="SteadyHands/climate-fallacy-roberta",
+    top_k=None,  # set top_k=3 to see top-3 fallacies
+)
 
-## Model Card Contact
+text = "Temperatures dropped this winter, so global warming must be a hoax."
+outputs = clf(text)
 
-[More Information Needed]
+print(outputs)