Online finetuning runs (sometimes crash because of nans)

README.md

@@ -15,12 +15,20 @@ conda activate lerobot
 python setup.py develop
 ```
 
+## TODO
+
+- [ ] priority update doesnt match FOWM or original paper
+- [ ] self.step=100000 should be updated at every step to adjust to horizon of planner
+- [ ] prefetch replay buffer to speedup training
+- [ ] parallelize env to speedup eval
 
 ## Contribute
 
 **style**
 ```
-isort .
-black .
+isort lerobot
+black lerobot
+isort test
+black test
 pylint lerobot
 ```