improve env_protection doc

deepmd/tf/descriptor/se_a.py

@@ -189,6 +189,8 @@ def __init__(
             raise RuntimeError(
                 f"rcut_smth ({rcut_smth:f}) should be no more than rcut ({rcut:f})!"
             )
+        if env_protection != 0.0:
+            raise NotImplementedError("env_protection != 0.0 is not supported.")
         self.sel_a = sel
         self.rcut_r = rcut
         self.rcut_r_smth = rcut_smth

deepmd/utils/argcheck.py

@@ -200,7 +200,7 @@ def descrpt_se_a_args():
     doc_trainable = "If the parameters in the embedding net is trainable"
     doc_seed = "Random seed for parameter initialization"
     doc_exclude_types = "The excluded pairs of types which have no interaction with each other. For example, `[[0, 1]]` means no interaction between type 0 and type 1."
-    doc_env_protection = "Protection parameter to prevent division by zero errors during environment matrix calculations."
+    doc_env_protection = "(Supported backend: PyTorch) Protection parameter to prevent division by zero errors during environment matrix calculations. For example, when using paddings, there may be zero distances of neighbors, which may make division by zero error during environment matrix calculations without protection."
     doc_set_davg_zero = "Set the normalization average to zero. This option should be set when `atom_ener` in the energy fitting is used"
 
     return [