Update docs about backend (#1287)

lululxvi · May 7, 2023 · cda8547 · cda8547
1 parent 4363618
commit cda8547
Showing 1 changed file with 4 additions and 4 deletions.
diff --git a/docs/user/installation.rst b/docs/user/installation.rst
@@ -58,7 +58,7 @@ DeepXDE supports TensorFlow 1.x (``tensorflow.compat.v1`` in TensorFlow 2.x), Te
 
 * Use the ``DDE_BACKEND`` environment variable:
 
-    - You can use ``DDE_BACKEND=BACKEND python pde.py`` to specify the backend:
+    - You can use ``DDE_BACKEND=BACKEND python pde.py`` to specify the backend. Currently ``BACKEND`` can be chosen from "tensorflow.compat.v1" (TensorFlow 1.x backend), "tensorflow" (TensorFlow 2.x backend), "pytorch" (PyTorch), "jax" (JAX), and "paddle" (PaddlePaddle).
 
     $ DDE_BACKEND=tensorflow.compat.v1 python pde.py
 
@@ -78,15 +78,15 @@ DeepXDE supports TensorFlow 1.x (``tensorflow.compat.v1`` in TensorFlow 2.x), Te
     - You can also use ``python -m deepxde.backend.set_default_backend BACKEND`` to set the default backend
     - In Windows, you can find ``config.json`` file under "C:/Users/Username/.deepxde" directory
 
-Currently ``BACKEND`` can be chosen from "tensorflow.compat.v1" (TensorFlow 1.x backend), "tensorflow" (TensorFlow 2.x backend), "pytorch" (PyTorch), "jax" (JAX), and "paddle" (PaddlePaddle). The default backend is TensorFlow 1.x.
+* If no backend is selected as above, DeepXDE will automatically find an available backend.
 
 Which backend should I choose?
 ``````````````````````````````
 
-Although TensorFlow 1.x is the default backend, it may not be your best choice. Here is a comparison between different backends:
+Here is a comparison between different backends:
 
 - Different backends support slightly different features, and switch to another backend if DeepXDE raised a backend-related error.
-    - Currently, the number of features supported is: TensorFlow 1.x > TensorFlow 2.x > PyTorch > JAX.
+    - Currently, the number of features supported is: PaddlePaddle ≈ TensorFlow 1.x > TensorFlow 2.x ≈ PyTorch > JAX.
     - Some features can be implemented easily (basically translating from one framework to another), and we welcome your contributions.
 - Different backends have different computational speed, and switch to another backend if the speed is an issue in your case.
     - We find that there is no backend that is always faster than the others.