Update README.md

README.md

@@ -1,3 +1,44 @@
----
-license: cc-by-nc-4.0
----
+---
+license: cc-by-nc-4.0
+language:
+- es
+base_model:
+- pysentimiento/robertuito-base-uncased
+datasets:
+- manueltonneau/spanish-hate-speech-superset
+tags:
+- RoBERTuito
+- hate_speech
+- misogyny
+- RoBERTa
+pipeline_tag: fill-mask
+library_name: transformers
+widget:
+  - text: "Ella es una<mask>"
+---
+# misoRoBERTuito-4e
+
+misoRoBERTuito-4e is a domain adaptation of [RoBERTuito](https://huggingface.co/pysentimiento/robertuito-base-uncased) language model, specifically adapted to the misogyny domain.
+
+It was adapted using a guided lexical masking strategy during masked language model (MLM) pretraining. 
+Instead of randomly masking tokens, we prioritized masking words appearing in a [misogyny-specific lexicon](https://github.com/fmplaza/hate-speech-spanish-lexicons/blob/master/misogyny_lexicon.txt). 
+The base corpus used for domain adaptation was the [Spanish Hate Speech Superset](https://huggingface.co/datasets/manueltonneau/spanish-hate-speech-superset).
+
+
+For training the model we used a batch size of 8, with a learning rate of 2e-5. We trained the model for four epochs using a NVIDIA GeForce RTX 5090 GPU.
+
+## Usage
+
+```python
+from transformers import pipeline
+pipe = pipeline("fill-mask", model="citiusLTL/misoRoBERTuito-4e")
+text = pipe("Ella es una<mask>")
+print(text)
+```
+
+## Load model directly
+```python
+from transformers import AutoTokenizer, AutoModelForMaskedLM
+tokenizer = AutoTokenizer.from_pretrained("citiusLTL/misoRoBERTuito-4e")
+model = AutoModelForMaskedLM.from_pretrained("citiusLTL/misoRoBERTuito-4e")
+```