Update README.md

README.md

@@ -0,0 +1,53 @@
+---
+language: italian
+widget:
+- text: "Quando nacque D'Annunzio?"
+  context: "D'Annunzio nacque nel 1863"
+---
+
+# Italian Bert Base Uncased on Squad-it
+
+## Model description
+
+This model is the uncased base version of the italian BERT trained by `dbmdz`, which you may find at `dbmdz/bert-base-italian-uncased`.
+
+#### How to use
+
+```python
+from transformers import pipeline
+
+nlp = pipeline('question-answering', model='antoniocappiello/bert-base-italian-uncased-squad-it')
+
+# nlp(context="D'Annunzio nacque nel 1863", question="Quando nacque D'Annunzio?")
+# {'score': 0.9990354180335999, 'start': 22, 'end': 25, 'answer': '1863'}
+```
+
+## Training data
+
+It has been trained on the question answering task using [SQuAD-it](http://sag.art.uniroma2.it/demo-software/squadit/), derived from the original SQuAD dataset and obtained through the semi-automatic translation of the SQuAD dataset in Italian.
+
+## Training procedure
+
+```bash
+python ./examples/run_squad.py \
+    --model_type bert \
+    --model_name_or_path ./bert-base-italian-uncased/ \
+    --do_train \
+    --do_eval \
+    --train_file ./squad_it_uncased/train-v1.1.json \
+    --predict_file ./squad_it_uncased/dev-v1.1.json \
+    --learning_rate 3e-5 \
+    --num_train_epochs 2 \
+    --max_seq_length 384 \
+    --doc_stride 128 \
+    --output_dir ./models/bert-base-italian-uncased-squad-it/ \
+    --per_gpu_eval_batch_size=3   \
+    --per_gpu_train_batch_size=3   \
+    --do_lower_case \
+```
+
+## Eval results
+
+| Model        | batch  | EM    | F1    |
+| ------------- |:-----:|:-----:| -----:|
+| BERT base uncased  |  3   |  **63.8**  | **75.30** |