convert decay_rate to stop_lr from old inputs

deepmd/utils/compat.py

@@ -5,6 +5,7 @@
 from pathlib import Path
 from typing import Any, Dict, Optional, Sequence, Union
 
+import numpy as np
 from deepmd.common import j_must_have
 
 
@@ -237,6 +238,25 @@ def _jcopy(src: Dict[str, Any], dst: Dict[str, Any], keys: Sequence[str]):
         dst[k] = src[k]
 
 
+def remove_decay_rate(jdata: Dict[str, Any]):
+    """convert decay_rate to stop_lr.
+
+    Parameters
+    ----------
+    jdata: Dict[str, Any]
+        input data
+    """
+    lr = jdata["learning_rate"]
+    if "decay_rate" in lr:
+        decay_rate = lr["decay_rate"]
+        start_lr = lr["start_lr"]
+        stop_step = jdata["training"]["stop_batch"]
+        decay_steps = lr["decay_steps"]
+        stop_lr = np.exp(np.log(decay_rate) * (stop_step / decay_steps)) * start_lr
+        lr["stop_lr"] = stop_lr
+        lr.pop("decay_rate")
+
+
 def convert_input_v1_v2(jdata: Dict[str, Any],
                         warning: bool = True,
                         dump: Optional[Union[str, Path]] = None) -> Dict[str, Any]:
@@ -259,6 +279,9 @@ def convert_input_v1_v2(jdata: Dict[str, Any],
 
     jdata["training"] = new_tr_cfg
 
+    # remove deprecated arguments
+    remove_decay_rate(jdata)
+
     if warning:
         _warning_input_v1_v2(dump)
     if dump is not None:

source/tests/compat_inputs/water_v2.json

@@ -17,7 +17,7 @@
     "learning_rate" :{
 		"type":		"exp",
 		"decay_steps":	5000,
-		"decay_rate":	0.95,
+		"stop_lr":	3.505266624882874e-08,
 		"start_lr":	0.001
     },