Update README.md

README.md

@@ -1,3 +1,34 @@
 ---
 license: cc-by-nc-sa-4.0
+language:
+- de
 ---
+
+## Model description
+This model is a fine-tuned version of the [BERT-base-german-cased model by deepset](https://huggingface.co/transformers/model_doc/distilbert.html) to classify German-language toxic comments. 
+
+## How to use
+
+You can use the model with the following code.
+
+```python
+from transformers import AutoModelForSequenceClassification, AutoTokenizer, TextClassificationPipeline
+
+model_path = "ankekat1000/toxic-bert-german"
+tokenizer = AutoTokenizer.from_pretrained(model_path)
+model = AutoModelForSequenceClassification.from_pretrained(model_path)
+
+pipeline =  TextClassificationPipeline(model=model, tokenizer=tokenizer)
+print(pipeline('This is a test text.'))
+```
+
+
+## Training data
+comming soone
+
+## Training procedure
+
+coming soon
+## Evaluation results
+
+coming soon