update

README.md

@@ -1,2 +1,35 @@
 # Calibrated-BERT-Fine-Tuning
-Code for Paper: Calibrated Pre-trained Language Model Fine-Tuning via Manifold Smoothing
+This repo contains our code for paper:
+
+Calibrated Language Model Fine-Tuning for In- and Out-of-Distribution Data, EMNLP2020.
+
+[[paper](https://arxiv.org/abs/2010.11506)]
+
+![SDE-Net](figure/illustration.png)
+
+## Training & Evaluation
+
+Training with BERT base:
+```
+CUDA_VISIBLE_DEVICES=0 python bert.py --dataset 20news-15 --seed 0
+```
+Evaluation with BERT base
+```
+python test.py --model base --in_dataset 20news-15 --out_dataset 20news-5 --index 0
+```
+Evaluation with temperature scaling
+```
+python test.py --model temperature --in_dataset 20news-15 --out_dataset 20news-5 --index 0
+```
+Evaluation with monte-carlo dropout
+```
+python test.py --model mc-dropout --in_dataset 20news-15 --out_dataset 20news-5 --eva_iter 10 --index 0
+```
+Training with manifold smoothing
+```
+CUDA_VISIBLE_DEVICES=0,1 python manifold-smoothing.py --dataset 20news-15 --seed 0 --eps_in 0.0001 --eps_out 0.001 --eps_y 0.1
+```
+Evaluation with manifold smoothing
+```
+python test.py --model manifold-smoothing --in_dataset 20news-15 --out_dataset 20news-5 --eps_in 0.0001 --eps_out 0.001 --eps_y 0.1
+```