diff --git a/docs/img/hBN_band_condband.png b/docs/img/hBN_band_condband.png
index c05888bc..69efcb73 100644
Binary files a/docs/img/hBN_band_condband.png and b/docs/img/hBN_band_condband.png differ
diff --git a/docs/img/hBN_band_first.png b/docs/img/hBN_band_first.png
index bb7219eb..8f1c5243 100644
Binary files a/docs/img/hBN_band_first.png and b/docs/img/hBN_band_first.png differ
diff --git a/docs/img/hBN_band_longtrain.png b/docs/img/hBN_band_longtrain.png
index 0c315a89..a9616c3b 100644
Binary files a/docs/img/hBN_band_longtrain.png and b/docs/img/hBN_band_longtrain.png differ
diff --git a/docs/img/hBN_band_pushrs.png b/docs/img/hBN_band_pushrs.png
new file mode 100644
index 00000000..eaf9ccf5
Binary files /dev/null and b/docs/img/hBN_band_pushrs.png differ
diff --git a/docs/img/hBN_band_pushw.png b/docs/img/hBN_band_pushw.png
new file mode 100644
index 00000000..8a0ff17c
Binary files /dev/null and b/docs/img/hBN_band_pushw.png differ
diff --git a/docs/img/hBN_band_strain.png b/docs/img/hBN_band_strain.png
index 11e6ca50..15b73a99 100644
Binary files a/docs/img/hBN_band_strain.png and b/docs/img/hBN_band_strain.png differ
diff --git a/docs/img/hBN_band_varycutoff.png b/docs/img/hBN_band_varycutoff.png
deleted file mode 100644
index 9f858ddf..00000000
Binary files a/docs/img/hBN_band_varycutoff.png and /dev/null differ
diff --git a/docs/quick_start/hands_on.md b/docs/quick_start/hands_on.md
index 778450ca..2807c43b 100644
--- a/docs/quick_start/hands_on.md
+++ b/docs/quick_start/hands_on.md
@@ -65,7 +65,9 @@ Having the data file and input parameter, we can start training our first **DeeP
     "nnsk": {
             "onsite": {"method": "none"},
             "hopping": {"method": "powerlaw", "rs":1.6, "w":0.3},
-            "freeze": false
+            "soc":{},
+            "freeze": false,
+            "push":false
         }
     }
 ```
@@ -77,8 +79,8 @@ Since we are using only the valence orbitals at this stage, we can limit the ene
     "bandinfo": {
         "band_min": 0,
         "band_max": 6,
-        "emin": -0.1,
-        "emax": 20.0
+        "emin": null,
+        "emax": null
     }
 ```
 
@@ -86,7 +88,7 @@ Using the follwing command and we can train the first model:
 
 ```bash
 cd deeptb/examples/hBN
-dptb train input_short.json -o ./first
+dptb train ./input/input_first.json -o ./first
 ```
 
 Here ``-o`` indicate the output directory. During the fitting procedure, we can see the loss curve of hBN is decrease consistently. When finished, we get the fitting results in folders ```first```.
@@ -95,6 +97,10 @@ By modify the checkpoint path in the script `plot_band.py` and running it, the b
 ```bash
 python plot_band.py
 ```
+or just using the command line 
+```bash
+dptb run ./run/band.json -i ./first/checkpoint/nnsk.best.pth -o ./band_plot
+```
 > Note: the ```basis``` setting in the plotting script must be the same as in the input.
 
 ![band_first](../img/hBN_band_first.png)
@@ -121,7 +127,7 @@ To train the conduction band, the energy window we previously set in `info.json`
 
 We can then start the training using the previous model and modified input:
 ```bash
-dptb train input_short.json -i ./first/checkpoint/nnsk.ep1000.pth -o ./condband
+dptb train input/input_condband.json -i ./first/checkpoint/nnsk.ep500.pth -o ./condband
 ```
 ``-i`` states initialize the model from the checkpoint file, where the previous model is provided.
 
@@ -147,7 +153,7 @@ We can further improve the accuracy by incorporating more features of our code,
 After setting we can run the training for strain model:
 
 ```bash
-dptb train input_short.json  -i ./condband/checkpoint/nnsk.ep500.pth -o ./strain
+dptb train input/input_strain.json  -i ./condband/checkpoint/nnsk.ep500.pth -o ./strain
 ```
 
 We can also plot the band structure of the strain model:
@@ -161,30 +167,74 @@ It looks ok, we can further improve the accuracy by adding more neighbours, and
         "nnsk": {
             "onsite": {"method": "strain", "rs":1.6, "w":0.3},
             "hopping": {"method": "powerlaw", "rs":1.6, "w": 0.3},
+            "soc":{},
             "push": {"rs_thr": 0.02, "period": 10},
             "freeze": false
         }
     }
 ```
-This means that we gradually add up the `rs` in decay function, pushing up to 3rd nearest neighbour for considering in calculating bonding. see the input file `hBN/input/3_varycutoff.json` for detail. Then we can run the training again:
+This means that we gradually add up the `rs` in decay function, pushing up to 3rd nearest neighbour for considering in calculating bonding. see the input file `hBN/input/input_push_rs.json` for detail. Then we can run the training again:
 
 ```bash
-dptb train input_short.json  -i ./strain/checkpoint/nnsk.ep500.pth -o ./varycutoff
+dptb train input/input_push_rs.json  -i ./strain/checkpoint/nnsk.ep500.pth -o ./push_rs
 ```
 
 We finally get the model with more neighbors. We can plot the result again:
 
-![band_varycutoff](../img/hBN_band_varycutoff.png)
+![band_varycutoff](../img/hBN_band_pushrs.png)
+
+
+we can further push the decay w to 0.2 and train the model again. modify the model options:
+```json
+    "model_options": {
+        "nnsk": {
+            "onsite": {"method": "strain", "rs":1.6, "w":0.3},
+            "hopping": {"method": "powerlaw", "rs":3.4, "w": 0.3},
+            "soc":{},
+            "push": {"w_thr": -0.001, "period": 10},
+            "freeze": false
+        }
+    }
+```
+note:  we change the hopping cutoff `rs` to 3.4, and the push w_thr to -0.001.
+
+see the input file `hBN/input/input_push_w.json` and run the training:
+
+```bash
+dptb train input/input_push_w.json  -i ./push_rs/checkpoint/nnsk.iter_rs3.400_w0.300.pth -o ./push_w
+```
+
+We can the plot the band structure again:
+
+![band_varycutoff](../img/hBN_band_pushw.png)
+
+
+We can again increase more training epochs, using the pushed parameters and turn off push tag. see the input file `hBN/input/input_final.json` and run the training:
+
+```json
+    "model_options": {
+        "nnsk": {
+            "onsite": {"method": "strain", "rs":1.6, "w":0.3},
+            "hopping": {"method": "powerlaw", "rs":3.4, "w": 0.2},
+            "soc":{},
+            "push": false,
+            "freeze": false
+        }
+    }
+```
+
+```bash
+dptb train input/input_final.json  -i ./push_w/checkpoint/nnsk.iter_rs3.400_w0.210.pth -o ./final
+```
 
-We can again increase more training epochs, using the larger cutoff checkpoint. This can be done simply by assigning a large `num_epoch` in `train_options`. 
 
 And we can get a fairly good fitting result:
 
 ![band_longtrain](../img/hBN_band_longtrain.png)
 
 Now you have learned the basis use of **DeePTB**, however, the advanced functions still need to be explored for accurate and flexible electron structure representation, such as:
-- atomic orbitals
 - environmental correction
+- spin-orbital interaction
 - ...
 
 Altogether, we can simulate the electronic structure of a crystal system in a dynamic trajectory. **DeePTB** is capable of handling atom movement, volume change under stress, SOC effect and can use DFT eigenvalues with different orbitals and xc functionals as training targets.
\ No newline at end of file
diff --git a/examples/hBN/band_plot.ipynb b/examples/hBN/band_plot.ipynb
index c798bc32..f45983ad 100644
--- a/examples/hBN/band_plot.ipynb
+++ b/examples/hBN/band_plot.ipynb
@@ -2,24 +2,25 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": 3,
    "metadata": {},
    "outputs": [],
    "source": [
     "from dptb.postprocess.bandstructure.band import Band\n",
     "from dptb.nn.nnsk import NNSK\n",
     "from dptb.utils.tools import j_loader\n",
+    "from dptb.nn.build import build_model\n",
     "import os"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": 4,
    "metadata": {},
    "outputs": [
     {
      "data": {
-      "image/png": "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",
+      "image/png": "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",
       "text/plain": [
        "<Figure size 450x400 with 1 Axes>"
       ]
@@ -30,8 +31,7 @@
    ],
    "source": [
     "\n",
-    "model = NNSK.from_reference(checkpoint=\"./ref_ckpts/condband/checkpoint/nnsk.ep500.pth\",\n",
-    "                            basis={'B': ['2s', '2p', 'd*'], 'N': ['2s', '2p', 'd*']})\n",
+    "model = build_model(checkpoint=\"./ref_ckpts/condband/checkpoint/nnsk.ep500.pth\")\n",
     "jdata = j_loader(\"./run/band.json\")\n",
     "results_path = \"./band_plot\"\n",
     "kpath_kwargs = jdata[\"task_options\"]\n",
diff --git a/examples/hBN/band_plot.py b/examples/hBN/band_plot.py
index 029efd5c..b71952d6 100644
--- a/examples/hBN/band_plot.py
+++ b/examples/hBN/band_plot.py
@@ -1,10 +1,11 @@
 from dptb.postprocess.bandstructure.band import Band
 from dptb.nn.nnsk import NNSK
 from dptb.utils.tools import j_loader
+from dptb.nn.build import build_model
 import os
 
-model = NNSK.from_reference(checkpoint="./ref_ckpts/condband/checkpoint/nnsk.ep500.pth",
-                            basis={'B': ['2s', '2p', 'd*'], 'N': ['2s', '2p', 'd*']})
+model = build_model(checkpoint="./ref_ckpts/condband/checkpoint/nnsk.ep500.pth")
+
 jdata = j_loader("./run/band.json")
 results_path = "./band_plot"
 kpath_kwargs = jdata["task_options"]
diff --git a/examples/hBN/band_plot/1.png b/examples/hBN/band_plot/1.png
deleted file mode 100644
index bb7219eb..00000000
Binary files a/examples/hBN/band_plot/1.png and /dev/null differ
diff --git a/examples/hBN/band_plot/2.png b/examples/hBN/band_plot/2.png
deleted file mode 100644
index c05888bc..00000000
Binary files a/examples/hBN/band_plot/2.png and /dev/null differ
diff --git a/examples/hBN/band_plot/3.png b/examples/hBN/band_plot/3.png
deleted file mode 100644
index 11e6ca50..00000000
Binary files a/examples/hBN/band_plot/3.png and /dev/null differ
diff --git a/examples/hBN/band_plot/4.png b/examples/hBN/band_plot/4.png
deleted file mode 100644
index 9f858ddf..00000000
Binary files a/examples/hBN/band_plot/4.png and /dev/null differ
diff --git a/examples/hBN/band_plot/5.png b/examples/hBN/band_plot/5.png
deleted file mode 100644
index 0c315a89..00000000
Binary files a/examples/hBN/band_plot/5.png and /dev/null differ
diff --git a/examples/hBN/band_plot/condband.png b/examples/hBN/band_plot/condband.png
new file mode 100644
index 00000000..69efcb73
Binary files /dev/null and b/examples/hBN/band_plot/condband.png differ
diff --git a/examples/hBN/band_plot/final.png b/examples/hBN/band_plot/final.png
new file mode 100644
index 00000000..a9616c3b
Binary files /dev/null and b/examples/hBN/band_plot/final.png differ
diff --git a/examples/hBN/band_plot/first.png b/examples/hBN/band_plot/first.png
new file mode 100644
index 00000000..8f1c5243
Binary files /dev/null and b/examples/hBN/band_plot/first.png differ
diff --git a/examples/hBN/band_plot/push_rs.png b/examples/hBN/band_plot/push_rs.png
new file mode 100644
index 00000000..eaf9ccf5
Binary files /dev/null and b/examples/hBN/band_plot/push_rs.png differ
diff --git a/examples/hBN/band_plot/push_w.png b/examples/hBN/band_plot/push_w.png
new file mode 100644
index 00000000..8a0ff17c
Binary files /dev/null and b/examples/hBN/band_plot/push_w.png differ
diff --git a/examples/hBN/band_plot/strain.png b/examples/hBN/band_plot/strain.png
new file mode 100644
index 00000000..15b73a99
Binary files /dev/null and b/examples/hBN/band_plot/strain.png differ
diff --git a/examples/hBN/data/kpath.0/info.json b/examples/hBN/data/kpath.0/info.json
index f087543b..b48038b6 100644
--- a/examples/hBN/data/kpath.0/info.json
+++ b/examples/hBN/data/kpath.0/info.json
@@ -11,7 +11,7 @@
     "bandinfo": {
         "band_min": 0,
         "band_max": 6,
-        "emin": -0.1,
-        "emax": 40.0
+        "emin": null,
+        "emax": null
     }
-}
\ No newline at end of file
+}
diff --git a/examples/hBN/input/2_condband.json b/examples/hBN/input/input_condband.json
similarity index 81%
rename from examples/hBN/input/2_condband.json
rename to examples/hBN/input/input_condband.json
index 81c9e84a..68e78449 100644
--- a/examples/hBN/input/2_condband.json
+++ b/examples/hBN/input/input_condband.json
@@ -1,44 +1,46 @@
-{
-    "common_options": {
-            "basis": {
-                "B": ["2s", "2p", "d*"],
-                "N": ["2s", "2p", "d*"]
-            },
-            "device": "cpu",
-            "dtype": "float32",
-            "overlap": false,
-            "seed": 114514
-    },
-    "train_options": {
-        "num_epoch": 500,
-        "batch_size": 1,
-        "optimizer": {
-            "lr": 0.05,
-            "type": "Adam"
-        },
-        "lr_scheduler": {
-            "type": "exp",
-            "gamma": 0.999
-        },
-        "loss_options":{
-            "train": {"method": "eigvals"}
-        },
-        "save_freq": 100,
-        "validation_freq": 10,
-        "display_freq": 10
-    },
-    "model_options": {
-        "nnsk": {
-            "onsite": {"method": "none"},
-            "hopping": {"method": "powerlaw", "rs":1.6, "w": 0.3},
-            "freeze": false
-        }
-    },
-    "data_options": {
-        "train": {
-            "root": "./data/",
-            "prefix": "kpath",
-            "get_eigenvalues": true
-        }
-    }
+{
+    "common_options": {
+            "basis": {
+                "B": ["2s", "2p","d*"],
+                "N": ["2s", "2p","d*"]
+            },
+            "device": "cpu",
+            "dtype": "float32",
+            "overlap": false,
+            "seed": 42
+    },
+    "train_options": {
+        "num_epoch": 500,
+        "batch_size": 1,
+        "optimizer": {
+            "lr": 0.05,
+            "type": "Adam"
+        },
+        "lr_scheduler": {
+            "type": "exp",
+            "gamma": 0.999
+        },
+        "loss_options":{
+            "train": {"method": "eigvals"}
+        },
+        "save_freq": 50,
+        "validation_freq": 10,
+        "display_freq": 10
+    },
+    "model_options": {
+        "nnsk": {
+            "onsite": {"method": "none"},
+            "hopping": {"method": "powerlaw", "rs":1.6, "w": 0.3},
+            "soc":{},
+            "freeze": false,
+            "push":false
+        }
+    },
+    "data_options": {
+        "train": {
+            "root": "./data/",
+            "prefix": "kpath",
+            "get_eigenvalues": true
+        }
+    }
 }
\ No newline at end of file
diff --git a/examples/hBN/input/5_longtrain.json b/examples/hBN/input/input_final.json
similarity index 67%
rename from examples/hBN/input/5_longtrain.json
rename to examples/hBN/input/input_final.json
index a78516d7..19c1edb2 100644
--- a/examples/hBN/input/5_longtrain.json
+++ b/examples/hBN/input/input_final.json
@@ -1,44 +1,46 @@
-{
-    "common_options": {
-            "basis": {
-                "B": ["2s", "2p", "d*"],
-                "N": ["2s", "2p", "d*"]
-            },
-            "device": "cpu",
-            "dtype": "float32",
-            "overlap": false,
-            "seed": 42
-    },
-    "train_options": {
-        "num_epoch": 10000,
-        "batch_size": 1,
-        "optimizer": {
-            "lr": 0.001,
-            "type": "Adam"
-        },
-        "lr_scheduler": {
-            "type": "exp",
-            "gamma": 0.999
-        },
-        "loss_options":{
-            "train": {"method": "eigvals"}
-        },
-        "save_freq": 1000,
-        "validation_freq": 10,
-        "display_freq": 10
-    },
-    "model_options": {
-        "nnsk": {
-            "onsite": {"method": "strain", "rs":1.6, "w":0.3},
-            "hopping": {"method": "powerlaw", "rs":3.6, "w": 0.3},
-            "freeze": false
-        }
-    },
-    "data_options": {
-        "train": {
-            "root": "./data/",
-            "prefix": "kpath",
-            "get_eigenvalues": true
-        }
-    }
+{
+    "common_options": {
+            "basis": {
+                "B": ["2s", "2p","d*"],
+                "N": ["2s", "2p","d*"]
+            },
+            "device": "cpu",
+            "dtype": "float32",
+            "overlap": false,
+            "seed": 42
+    },
+    "train_options": {
+        "num_epoch": 1000,
+        "batch_size": 1,
+        "optimizer": {
+            "lr": 0.05,
+            "type": "Adam"
+        },
+        "lr_scheduler": {
+            "type": "exp",
+            "gamma": 0.999
+        },
+        "loss_options":{
+            "train": {"method": "eigvals"}
+        },
+        "save_freq": 10,
+        "validation_freq": 10,
+        "display_freq": 10
+    },
+    "model_options": {
+        "nnsk": {
+            "onsite": {"method": "strain", "rs": 1.6, "w": 0.3},
+            "hopping": {"method": "powerlaw", "rs":3.4, "w": 0.2},
+            "soc":{},
+            "freeze": false,
+            "push": false
+        }
+    },
+    "data_options": {
+        "train": {
+            "root": "./data/",
+            "prefix": "kpath",
+            "get_eigenvalues": true
+        }
+    }
 }
\ No newline at end of file
diff --git a/examples/hBN/input/1_start.json b/examples/hBN/input/input_first.json
similarity index 83%
rename from examples/hBN/input/1_start.json
rename to examples/hBN/input/input_first.json
index 98f771c5..ce1f0330 100644
--- a/examples/hBN/input/1_start.json
+++ b/examples/hBN/input/input_first.json
@@ -1,44 +1,46 @@
-{
-    "common_options": {
-            "basis": {
-                "B": ["2s", "2p"],
-                "N": ["2s", "2p"]
-            },
-            "device": "cpu",
-            "dtype": "float32",
-            "overlap": false,
-            "seed": 114514
-    },
-    "train_options": {
-        "num_epoch": 1000,
-        "batch_size": 1,
-        "optimizer": {
-            "lr": 0.02,
-            "type": "Adam"
-        },
-        "lr_scheduler": {
-            "type": "exp",
-            "gamma": 0.999
-        },
-        "loss_options":{
-            "train": {"method": "eigvals"}
-        },
-        "save_freq": 100,
-        "validation_freq": 10,
-        "display_freq": 10
-    },
-    "model_options": {
-        "nnsk": {
-            "onsite": {"method": "none"},
-            "hopping": {"method": "powerlaw", "rs":1.6, "w": 0.3},
-            "freeze": false
-        }
-    },
-    "data_options": {
-        "train": {
-            "root": "./data/",
-            "prefix": "kpath",
-            "get_eigenvalues": true
-        }
-    }
+{
+    "common_options": {
+            "basis": {
+                "B": ["2s", "2p"],
+                "N": ["2s", "2p"]
+            },
+            "device": "cpu",
+            "dtype": "float32",
+            "overlap": false,
+            "seed": 42
+    },
+    "train_options": {
+        "num_epoch": 500,
+        "batch_size": 1,
+        "optimizer": {
+            "lr": 0.05,
+            "type": "Adam"
+        },
+        "lr_scheduler": {
+            "type": "exp",
+            "gamma": 0.999
+        },
+        "loss_options":{
+            "train": {"method": "eigvals"}
+        },
+        "save_freq": 50,
+        "validation_freq": 10,
+        "display_freq": 10
+    },
+    "model_options": {
+        "nnsk": {
+            "onsite": {"method": "none"},
+            "hopping": {"method": "powerlaw", "rs":1.6, "w": 0.3},
+            "soc":{},
+            "freeze": false,
+            "push":false
+        }
+    },
+    "data_options": {
+        "train": {
+            "root": "./data/",
+            "prefix": "kpath",
+            "get_eigenvalues": true
+        }
+    }
 }
\ No newline at end of file
diff --git a/examples/hBN/input/4_varycutoff.json b/examples/hBN/input/input_push_rs.json
similarity index 69%
rename from examples/hBN/input/4_varycutoff.json
rename to examples/hBN/input/input_push_rs.json
index 464c2ea1..4647b6fa 100644
--- a/examples/hBN/input/4_varycutoff.json
+++ b/examples/hBN/input/input_push_rs.json
@@ -1,19 +1,19 @@
 {
     "common_options": {
             "basis": {
-                "B": ["2s", "2p", "d*"],
-                "N": ["2s", "2p", "d*"]
+                "B": ["2s", "2p","d*"],
+                "N": ["2s", "2p","d*"]
             },
             "device": "cpu",
             "dtype": "float32",
             "overlap": false,
-            "seed": 114514
+            "seed": 42
     },
     "train_options": {
-        "num_epoch": 1100,
+        "num_epoch": 1000,
         "batch_size": 1,
         "optimizer": {
-            "lr": 0.01,
+            "lr": 0.05,
             "type": "Adam"
         },
         "lr_scheduler": {
@@ -23,16 +23,17 @@
         "loss_options":{
             "train": {"method": "eigvals"}
         },
-        "save_freq": 100,
+        "save_freq": 50,
         "validation_freq": 10,
         "display_freq": 10
     },
     "model_options": {
         "nnsk": {
-            "onsite": {"method": "strain", "rs":1.6, "w":0.3},
+            "onsite": {"method": "strain", "rs": 1.6, "w": 0.3},
             "hopping": {"method": "powerlaw", "rs":1.6, "w": 0.3},
-            "push": {"rs_thr": 0.02, "period": 10},
-            "freeze": false
+            "soc":{},
+            "freeze": false,
+            "push": {"rs_thr": 0.02, "period": 10}
         }
     },
     "data_options": {
diff --git a/examples/hBN/input/input_push_w.json b/examples/hBN/input/input_push_w.json
new file mode 100644
index 00000000..d2e89098
--- /dev/null
+++ b/examples/hBN/input/input_push_w.json
@@ -0,0 +1,46 @@
+{
+    "common_options": {
+            "basis": {
+                "B": ["2s", "2p","d*"],
+                "N": ["2s", "2p","d*"]
+            },
+            "device": "cpu",
+            "dtype": "float32",
+            "overlap": false,
+            "seed": 42
+    },
+    "train_options": {
+        "num_epoch": 1000,
+        "batch_size": 1,
+        "optimizer": {
+            "lr": 0.05,
+            "type": "Adam"
+        },
+        "lr_scheduler": {
+            "type": "exp",
+            "gamma": 0.999
+        },
+        "loss_options":{
+            "train": {"method": "eigvals"}
+        },
+        "save_freq": 10,
+        "validation_freq": 10,
+        "display_freq": 10
+    },
+    "model_options": {
+        "nnsk": {
+            "onsite": {"method": "strain", "rs": 1.6, "w": 0.3},
+            "hopping": {"method": "powerlaw", "rs":3.4, "w": 0.3},
+            "soc":{},
+            "freeze": false,
+            "push": {"w_thr": -0.001, "period": 10}
+        }
+    },
+    "data_options": {
+        "train": {
+            "root": "./data/",
+            "prefix": "kpath",
+            "get_eigenvalues": true
+        }
+    }
+}
\ No newline at end of file
diff --git a/examples/hBN/input/3_strain.json b/examples/hBN/input/input_strain.json
similarity index 73%
rename from examples/hBN/input/3_strain.json
rename to examples/hBN/input/input_strain.json
index 464972f9..c8683f34 100644
--- a/examples/hBN/input/3_strain.json
+++ b/examples/hBN/input/input_strain.json
@@ -1,44 +1,46 @@
-{
-    "common_options": {
-            "basis": {
-                "B": ["2s", "2p", "d*"],
-                "N": ["2s", "2p", "d*"]
-            },
-            "device": "cpu",
-            "dtype": "float32",
-            "overlap": false,
-            "seed": 114514
-    },
-    "train_options": {
-        "num_epoch": 500,
-        "batch_size": 1,
-        "optimizer": {
-            "lr": 0.01,
-            "type": "Adam"
-        },
-        "lr_scheduler": {
-            "type": "exp",
-            "gamma": 0.999
-        },
-        "loss_options":{
-            "train": {"method": "eigvals"}
-        },
-        "save_freq": 100,
-        "validation_freq": 10,
-        "display_freq": 10
-    },
-    "model_options": {
-        "nnsk": {
-            "onsite": {"method": "strain", "rs":1.6, "w":0.3},
-            "hopping": {"method": "powerlaw", "rs":1.6, "w": 0.3},
-            "freeze": false
-        }
-    },
-    "data_options": {
-        "train": {
-            "root": "./data/",
-            "prefix": "kpath",
-            "get_eigenvalues": true
-        }
-    }
+{
+    "common_options": {
+            "basis": {
+                "B": ["2s", "2p","d*"],
+                "N": ["2s", "2p","d*"]
+            },
+            "device": "cpu",
+            "dtype": "float32",
+            "overlap": false,
+            "seed": 42
+    },
+    "train_options": {
+        "num_epoch": 500,
+        "batch_size": 1,
+        "optimizer": {
+            "lr": 0.05,
+            "type": "Adam"
+        },
+        "lr_scheduler": {
+            "type": "exp",
+            "gamma": 0.999
+        },
+        "loss_options":{
+            "train": {"method": "eigvals"}
+        },
+        "save_freq": 50,
+        "validation_freq": 10,
+        "display_freq": 10
+    },
+    "model_options": {
+        "nnsk": {
+            "onsite": {"method": "strain", "rs": 1.6, "w": 0.3},
+            "hopping": {"method": "powerlaw", "rs":1.6, "w": 0.3},
+            "soc":{},
+            "freeze": false,
+            "push":false
+        }
+    },
+    "data_options": {
+        "train": {
+            "root": "./data/",
+            "prefix": "kpath",
+            "get_eigenvalues": true
+        }
+    }
 }
\ No newline at end of file
diff --git a/examples/hBN/input_short.json b/examples/hBN/input_short.json
index 97bf6192..ce1f0330 100644
--- a/examples/hBN/input_short.json
+++ b/examples/hBN/input_short.json
@@ -23,7 +23,7 @@
         "loss_options":{
             "train": {"method": "eigvals"}
         },
-        "save_freq": 100,
+        "save_freq": 50,
         "validation_freq": 10,
         "display_freq": 10
     },
@@ -31,7 +31,9 @@
         "nnsk": {
             "onsite": {"method": "none"},
             "hopping": {"method": "powerlaw", "rs":1.6, "w": 0.3},
-            "freeze": false
+            "soc":{},
+            "freeze": false,
+            "push":false
         }
     },
     "data_options": {
diff --git a/examples/hBN/ref_ckpts/condband/checkpoint/nnsk.ep500.pth b/examples/hBN/ref_ckpts/condband/checkpoint/nnsk.ep500.pth
index b4f49222..ec6ff436 100644
Binary files a/examples/hBN/ref_ckpts/condband/checkpoint/nnsk.ep500.pth and b/examples/hBN/ref_ckpts/condband/checkpoint/nnsk.ep500.pth differ
diff --git a/examples/hBN/ref_ckpts/condband/checkpoint/nnsk.iter500.pth b/examples/hBN/ref_ckpts/condband/checkpoint/nnsk.iter500.pth
index 7195dc65..0b2ee078 100644
Binary files a/examples/hBN/ref_ckpts/condband/checkpoint/nnsk.iter500.pth and b/examples/hBN/ref_ckpts/condband/checkpoint/nnsk.iter500.pth differ
diff --git a/examples/hBN/ref_ckpts/final/checkpoint/nnsk.ep1000.pth b/examples/hBN/ref_ckpts/final/checkpoint/nnsk.ep1000.pth
new file mode 100644
index 00000000..7224baa0
Binary files /dev/null and b/examples/hBN/ref_ckpts/final/checkpoint/nnsk.ep1000.pth differ
diff --git a/examples/hBN/ref_ckpts/final/checkpoint/nnsk.iter1000.pth b/examples/hBN/ref_ckpts/final/checkpoint/nnsk.iter1000.pth
new file mode 100644
index 00000000..c0f04f6f
Binary files /dev/null and b/examples/hBN/ref_ckpts/final/checkpoint/nnsk.iter1000.pth differ
diff --git a/examples/hBN/ref_ckpts/first/checkpoint/nnsk.ep1000.pth b/examples/hBN/ref_ckpts/first/checkpoint/nnsk.ep1000.pth
deleted file mode 100644
index 246b4d73..00000000
Binary files a/examples/hBN/ref_ckpts/first/checkpoint/nnsk.ep1000.pth and /dev/null differ
diff --git a/examples/hBN/ref_ckpts/first/checkpoint/nnsk.ep500.pth b/examples/hBN/ref_ckpts/first/checkpoint/nnsk.ep500.pth
new file mode 100644
index 00000000..693d9439
Binary files /dev/null and b/examples/hBN/ref_ckpts/first/checkpoint/nnsk.ep500.pth differ
diff --git a/examples/hBN/ref_ckpts/first/checkpoint/nnsk.iter1000.pth b/examples/hBN/ref_ckpts/first/checkpoint/nnsk.iter1000.pth
deleted file mode 100644
index 5243ffba..00000000
Binary files a/examples/hBN/ref_ckpts/first/checkpoint/nnsk.iter1000.pth and /dev/null differ
diff --git a/examples/hBN/ref_ckpts/first/checkpoint/nnsk.iter500.pth b/examples/hBN/ref_ckpts/first/checkpoint/nnsk.iter500.pth
new file mode 100644
index 00000000..a2cb7c1f
Binary files /dev/null and b/examples/hBN/ref_ckpts/first/checkpoint/nnsk.iter500.pth differ
diff --git a/examples/hBN/ref_ckpts/longtrain/checkpoint/nnsk.iter10000.pth b/examples/hBN/ref_ckpts/longtrain/checkpoint/nnsk.iter10000.pth
deleted file mode 100644
index 2513fdc9..00000000
Binary files a/examples/hBN/ref_ckpts/longtrain/checkpoint/nnsk.iter10000.pth and /dev/null differ
diff --git a/examples/hBN/ref_ckpts/push_rs/checkpoint/nnsk.iter_rs3.400_w0.300.pth b/examples/hBN/ref_ckpts/push_rs/checkpoint/nnsk.iter_rs3.400_w0.300.pth
new file mode 100644
index 00000000..f7fb6a8d
Binary files /dev/null and b/examples/hBN/ref_ckpts/push_rs/checkpoint/nnsk.iter_rs3.400_w0.300.pth differ
diff --git a/examples/hBN/ref_ckpts/push_w/checkpoint/nnsk.iter_rs3.400_w0.210.pth b/examples/hBN/ref_ckpts/push_w/checkpoint/nnsk.iter_rs3.400_w0.210.pth
new file mode 100644
index 00000000..5f50f062
Binary files /dev/null and b/examples/hBN/ref_ckpts/push_w/checkpoint/nnsk.iter_rs3.400_w0.210.pth differ
diff --git a/examples/hBN/ref_ckpts/strain/checkpoint/nnsk.ep500.pth b/examples/hBN/ref_ckpts/strain/checkpoint/nnsk.ep500.pth
index 790601a9..07e09b00 100644
Binary files a/examples/hBN/ref_ckpts/strain/checkpoint/nnsk.ep500.pth and b/examples/hBN/ref_ckpts/strain/checkpoint/nnsk.ep500.pth differ
diff --git a/examples/hBN/ref_ckpts/strain/checkpoint/nnsk.iter500.pth b/examples/hBN/ref_ckpts/strain/checkpoint/nnsk.iter500.pth
index 1a841e07..a64b9d2d 100644
Binary files a/examples/hBN/ref_ckpts/strain/checkpoint/nnsk.iter500.pth and b/examples/hBN/ref_ckpts/strain/checkpoint/nnsk.iter500.pth differ
diff --git a/examples/hBN/ref_ckpts/test/checkpoint/best_nnsk_b1.600_c1.600_w0.300.pth b/examples/hBN/ref_ckpts/test/checkpoint/best_nnsk_b1.600_c1.600_w0.300.pth
deleted file mode 100644
index b12dee05..00000000
Binary files a/examples/hBN/ref_ckpts/test/checkpoint/best_nnsk_b1.600_c1.600_w0.300.pth and /dev/null differ
diff --git a/examples/hBN/ref_ckpts/test/checkpoint/best_nnsk_b3.600_c1.600_w0.300.pth b/examples/hBN/ref_ckpts/test/checkpoint/best_nnsk_b3.600_c1.600_w0.300.pth
deleted file mode 100644
index 31d69956..00000000
Binary files a/examples/hBN/ref_ckpts/test/checkpoint/best_nnsk_b3.600_c1.600_w0.300.pth and /dev/null differ
diff --git a/examples/hBN/ref_ckpts/test/checkpoint/latest_nnsk_b1.600_c1.600_w0.300.pth b/examples/hBN/ref_ckpts/test/checkpoint/latest_nnsk_b1.600_c1.600_w0.300.pth
deleted file mode 100644
index b00f422d..00000000
Binary files a/examples/hBN/ref_ckpts/test/checkpoint/latest_nnsk_b1.600_c1.600_w0.300.pth and /dev/null differ
diff --git a/examples/hBN/ref_ckpts/varycutoff/checkpoint/nnsk.iter1100.pth b/examples/hBN/ref_ckpts/varycutoff/checkpoint/nnsk.iter1100.pth
deleted file mode 100644
index 8380ee75..00000000
Binary files a/examples/hBN/ref_ckpts/varycutoff/checkpoint/nnsk.iter1100.pth and /dev/null differ
diff --git a/examples/hBN/run/band.json b/examples/hBN/run/band.json
index 2ce76827..421ada3c 100644
--- a/examples/hBN/run/band.json
+++ b/examples/hBN/run/band.json
@@ -1,4 +1,5 @@
 {   
+    "structure":"./data/struct.vasp",
     "task_options": {
         "task": "band",
         "kline_type":"abacus",
@@ -8,11 +9,13 @@
             [0.3333333, 0.3333333, 0, 50],
             [0, 0, 0, 1]
         ],
-        "nkpoints":151,
+        "nel_atom":{"N":5,"B":3},
         "klabels":["G", "M", "K", "G"],
-        "E_fermi":-9.87,
-        "emin":-30,
-        "emax":5,
+        "E_fermi":-12.798759460449219,
+        "emin":-25,
+        "emax":15,
         "ref_band": "./data/kpath.0/eigenvalues.npy"
-    }
+    },
+    "AtomicData_options" : {"r_max": 5.0, "oer_max":1.6, "pbc": true}
+
 }
diff --git a/examples/silicon/ref_ckpts/dptb/checkpoint/mix.ep50.pth b/examples/silicon/ref_ckpts/dptb/checkpoint/mix.ep50.pth
index 8073bdfe..f6f1a41b 100644
Binary files a/examples/silicon/ref_ckpts/dptb/checkpoint/mix.ep50.pth and b/examples/silicon/ref_ckpts/dptb/checkpoint/mix.ep50.pth differ
diff --git a/examples/silicon/ref_ckpts/dptb/checkpoint/mix.iter500.pth b/examples/silicon/ref_ckpts/dptb/checkpoint/mix.iter500.pth
index 411a15c5..fb8173de 100644
Binary files a/examples/silicon/ref_ckpts/dptb/checkpoint/mix.iter500.pth and b/examples/silicon/ref_ckpts/dptb/checkpoint/mix.iter500.pth differ
diff --git a/examples/silicon/ref_ckpts/dptb/train_config.json b/examples/silicon/ref_ckpts/dptb/train_config.json
index 6e94772f..2908272f 100644
--- a/examples/silicon/ref_ckpts/dptb/train_config.json
+++ b/examples/silicon/ref_ckpts/dptb/train_config.json
@@ -110,6 +110,7 @@
             "prefix": "kpath_spk",
             "get_eigenvalues": true,
             "type": "DefaultDataset",
+            "separator": ".",
             "get_Hamiltonian": false,
             "get_overlap": false,
             "get_DM": false