add simplify example of MAPbI3-scan based on lebesgue

dpgen/generator/arginfo.py

@@ -218,6 +218,8 @@ def fp_style_vasp_args() -> List[Argument]:
     doc_fp_incar = 'Input file for VASP. INCAR must specify KSPACING and KGAMMA.'
     doc_fp_aniso_kspacing = 'Set anisotropic kspacing. Usually useful for 1-D or 2-D materials. Only support VASP. If it is setting the KSPACING key in INCAR will be ignored.'
     doc_cvasp = 'If cvasp is true, DP-GEN will use Custodian to help control VASP calculation.'
+    doc_ratio_failed = 'Check the ratio of unsuccessfully terminated jobs. If too many FP tasks are not converged, RuntimeError will be raised.'
+    doc_fp_skip_bad_box = 'Skip the configurations that are obviously unreasonable before 02.fp'
 
     return [
         Argument("fp_pp_path", str, optional=False, doc=doc_fp_pp_path),
@@ -226,6 +228,10 @@ def fp_style_vasp_args() -> List[Argument]:
         Argument("fp_aniso_kspacing", list, optional=True,
                  doc=doc_fp_aniso_kspacing),
         Argument("cvasp", bool, optional=True, doc=doc_cvasp),
+        Argument("ratio_failed", float, optional=True,
+                 doc=doc_ratio_failed),
+        Argument("fp_skip_bad_box", str, optional=True,
+                 doc=doc_fp_skip_bad_box),
     ]
 
 # abacus

dpgen/simplify/arginfo.py

@@ -25,6 +25,7 @@ def general_simplify_arginfo() -> Argument:
     doc_iter_pick_number = "The number of pick data in each iteration."
     doc_model_devi_f_trust_lo = "The lower bound of forces for the selection for the model deviation."
     doc_model_devi_f_trust_hi = "The higher bound of forces for the selection for the model deviation."
+    doc_shuffle_poscar = 'Shuffle atoms of each frame before running simulations. The purpose is to sample the element occupation of alloys.'
 
     return [
         Argument("labeled", bool, optional=True, default=False, doc=doc_labeled),
@@ -33,6 +34,7 @@ def general_simplify_arginfo() -> Argument:
         Argument("iter_pick_number", int, doc=doc_iter_pick_number),
         Argument("model_devi_f_trust_lo", float, optional=False, doc=doc_model_devi_f_trust_lo),
         Argument("model_devi_f_trust_hi", float, optional=False, doc=doc_model_devi_f_trust_hi),
+        Argument("shuffle_poscar", bool, optional=True, default=False, doc=doc_shuffle_poscar),
     ]
 
 
@@ -71,10 +73,16 @@ def fp_args() -> List[Argument]:
     """
     doc_fp_task_max = 'Maximum of structures to be calculated in 02.fp of each iteration.'
     doc_fp_task_min = 'Minimum of structures to be calculated in 02.fp of each iteration.'
+    doc_fp_accurate_threshold = 'If the accurate ratio is larger than this number, no fp calculation will be performed, i.e. fp_task_max = 0.'
+    doc_fp_accurate_soft_threshold = 'If the accurate ratio is between this number and fp_accurate_threshold, the fp_task_max linearly decays to zero.'
 
     return [
         Argument("fp_task_max", int, optional=True, doc=doc_fp_task_max),
         Argument("fp_task_min", int, optional=True, doc=doc_fp_task_min),
+        Argument("fp_accurate_threshold", float,
+                 optional=True, doc=doc_fp_accurate_threshold),
+        Argument("fp_accurate_soft_threshold", float,
+                 optional=True, doc=doc_fp_accurate_soft_threshold),
     ]
 
 

examples/simplify-MAPbI3-scan-lebesgue/README.md

@@ -0,0 +1,7 @@
+This is an example for `dpgen simplify`. `data` contains a simplistic data set based on MAPbI3-scan case. Since it has been greatly reduced, do not take it seriously. It is just a demo. 
+`simplify_example` is the work path, which contains `INCAR` and templates for `simplify.json` and `machine.json`. You can use the command `nohup dpgen simplify simplify.json machine.json 1>log 2>err &` here to test if `dpgen simplify` can run normally. 
+
+Kindly reminder: 
+1. `machine.json` is supported by `dpdispatcher 0.4.15`, please check https://docs.deepmodeling.com/projects/dpdispatcher/en/latest/ to update the parameters according to your `dpdispatcher` version.
+2. `POTCAR` should be prepared by the user. 
+3. Please check the path and files name and make sure they are correct. 

examples/simplify-MAPbI3-scan-lebesgue/data/sys.000/type.raw

@@ -0,0 +1,48 @@
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+1
+1
+1
+1
+2
+2
+2
+2
+3
+3
+3
+3
+4
+4
+4
+4
+4
+4
+4
+4
+4
+4
+4
+4
+4
+4
+4
+4
+4
+4
+4
+4
+4
+4
+4
+4

examples/simplify-MAPbI3-scan-lebesgue/data/sys.000/type_map.raw

@@ -0,0 +1,5 @@
+I
+Pb
+C
+N
+H

examples/simplify-MAPbI3-scan-lebesgue/simplify_example/INCAR

@@ -0,0 +1,38 @@
+#Parameters
+SYSTEM  =  dpgen_rlx
+PREC    = A
+ISTART  = 0
+ICHARG  = 2
+IWAVPR  =  11
+#Electronic Relaxation
+ENCUT  = 500
+NELM   = 100
+NELMIN = 6
+NELMDL = -5
+EDIFF  = 1e-06
+LREAL  = .False.
+ALGO   = Normal
+# GGA = PE
+METAGGA = SCAN
+#Ionic Relaxation
+IBRION = 2
+POTIM  = 0.2
+ISIF   = 3
+EDIFFG = -0.005
+ISYM   = 0
+NSW    = 0
+ISMEAR = 0
+SIGMA  = 0.1
+#vdW correction
+# IVDW = 12  #D3-BJ
+# Write flags
+LWAVE   = .False.
+LCHARG  = .False.
+
+KSPACING = 0.16
+KGAMMA   = .TRUE.
+#KPAR    = 1
+NPAR    = 2
+
+PSTRESS  = 0.0
+# NCORE    = 4

examples/simplify-MAPbI3-scan-lebesgue/simplify_example/machine.json

@@ -0,0 +1,112 @@
+{
+    "api_version": "1.0",
+    "_comment": "dependent on dpdispatcher 0.4.15",
+    "train" :[
+      {
+        "command": "dp",
+        "machine": {
+        "batch_type": "Lebesgue",
+        "context_type": "LebesgueContext",
+        "local_root" : "./",
+          "remote_profile":{
+          "email": "",
+          "password": "",
+          "program_id": ,
+              "input_data":{
+                  "api_version":2,
+                  "job_type": "indicate",
+                  "log_file": "00*/lcurve.out",
+                  "grouped": true,
+                  "job_name": "simplify_MAPbI3-scan_train_job",
+                  "disk_size": 100,
+                  "scass_type":"c8_m32_1 * NVIDIA V100",
+                  "platform": "ali",
+                  "checkpoint_files": ["00*/model.ckpt*", "00*/checkpoint"],
+                  "checkpoint_time": 15,
+                  "image_name":"LBG_DeePMD-kit_2.1.1_v1.1",
+                  "on_demand":0
+              }
+          }
+        },
+        "resources": {
+          "number_node": 15,
+          "cpu_per_node": 4,
+          "gpu_per_node": 1,
+          "queue_name": "V100_8_32",
+          "group_size": 1
+        }
+      }],
+    "model_devi":
+      [{
+        "command": "dp",
+        "machine": {
+        "batch_type": "Lebesgue",
+        "context_type": "LebesgueContext",
+        "local_root" : "./",
+          "remote_profile":{
+          "email": "",
+          "password": "",
+          "program_id": ,
+              "input_data":{
+                "api_version":2,
+                "job_type": "indicate",
+                "log_file": "cubic-*/*/md.log",
+                "grouped": true,
+                "job_name": "MAPbI3-scan_md_job",
+                "disk_size": 100,
+                "scass_type":"c8_m32_1 * NVIDIA V100",
+                "platform": "ali",
+                "checkpoint_files": "sync_files",
+                "checkpoint_time": 15,
+                "image_name":"LBG_DeePMD-kit_2.1.1_v1.1",
+                "on_demand":0
+              }
+          }
+        },
+        "resources": {
+          "number_node": 15,
+          "cpu_per_node": 4,
+          "gpu_per_node": 1,
+          "queue_name": "V100_8_32",
+          "group_size": 10
+        }
+      }],
+    "fp":
+      [{
+        "command": "ulimit -m unlimited; ulimit -s unlimited; mpirun -n 32 vasp_std",
+        "machine": {
+        "batch_type": "Lebesgue",
+        "context_type": "LebesgueContext",
+        "local_root" : "./",
+          "remote_profile":{
+          "email": "",
+          "password": "",
+          "program_id": ,
+              "input_data":{
+                "api_version":2,
+                "job_type": "indicate",
+                "log_file": "**/fp.log",
+                "grouped": true,
+                "job_name": "simplify_MAPbI3-scan_fp_job",
+                "disk_size": 100,
+                "scass_type":"c32_m64_cpu",
+                "platform": "ali",
+                "checkpoint_files": "sync_files",
+                "checkpoint_time": 15,
+                "image_name":"LBG_vasp-5-4-4_v1",
+                "on_demand":0
+              }
+          }
+        },
+        "resources": {
+          "number_node": 15,
+          "cpu_per_node": 32,
+          "gpu_per_node": 0,
+          "queue_name": "CPU",
+          "group_size": 10,
+      "source_list": ["/opt/intel/oneapi/setvars.sh"]
+        }
+      }
+    ]
+  }
+

examples/simplify-MAPbI3-scan-lebesgue/simplify_example/simplify.json

@@ -0,0 +1,95 @@
+{
+  "type_map": ["I","Pb","C","N","H"],
+  "mass_map": [126.90447, 207.2, 12.0108, 14.0067, 1.00795],
+
+  "pick_data":  "../data", 
+  "init_data_prefix": "",
+  "init_data_sys":   [],
+  "sys_configs": [null],
+  "sys_batch_size":  [1], 
+
+  "_comment": " 00.train ",
+  "numb_models": 4,
+  "model_devi_activation_func":[["tanh","tanh"],["tanh","gelu"],["gelu","gelu"],["gelu","tanh"]],
+
+  "default_training_param": {
+    "model": {
+      "type_map":           ["I","Pb","C","N","H"],
+      "descriptor": {        
+        "type":             "se_e2_a",
+        "sel": 
+        [
+          20,
+          8,
+          8,
+          8,
+          48
+        ],
+        "rcut_smth":        2.00,
+        "rcut":             6.00,
+        "neuron":           [25, 50, 100],
+        "resnet_dt":        false,
+        "type_one_side":     true,
+        "trainable":      true,
+	      "axis_neuron":      12,
+        "seed":             0
+      },
+      "fitting_net": {
+        "neuron":           [240, 240, 240],
+        "resnet_dt":        true,
+        "trainable":      [true, true, true, true],
+        "seed": 0
+      }
+    },
+      "loss": {
+        "start_pref_e": 0.02,
+        "limit_pref_e": 2,
+        "start_pref_f": 1000,
+        "limit_pref_f": 2,
+        "start_pref_v": 0.01,
+        "limit_pref_v": 1
+    },
+      "learning_rate": {
+        "type":              "exp",
+        "start_lr":          0.001,
+        "decay_steps":       25,
+        "decay_rate":        0.95
+    },
+      "training": {
+        "set_prefix":        "set",
+        "stop_batch":        5000,
+        "batch_size":        "auto",
+        "seed": 1,
+        "_comment":          "frequencies counted in batch",
+        "disp_file":         "lcurve.out",
+        "disp_freq":         1000,
+        "numb_test":         4,
+        "save_freq":         10000,
+        "save_ckpt":         "model.ckpt", 
+        "disp_training":     true,
+        "time_training":     true,
+        "profiling":         false,
+        "profiling_file":    "timeline.json"
+    }
+  },
+
+  "_comment":                   "02.fp",
+  "fp_style":                   "vasp",
+  "fp_skip_bad_box":  "length_ratio:5;height_ratio:5",
+  "fp_accurate_threshold":       0.95,
+  "fp_accurate_soft_threshold":  0.90,
+  "shuffle_poscar":              false,
+  "fp_task_max":                 20,
+  "fp_task_min":                 5,
+  "ratio_failed":                0.30,
+  "fp_pp_path": "../vasp_inputs/",
+  "fp_pp_files": ["POTCAR_I","POTCAR_Pb_d","POTCAR_C","POTCAR_N","POTCAR_H"],
+  "fp_incar": "INCAR",
+
+  "labeled": false,
+  "init_pick_number":5,
+  "iter_pick_number":5,
+  "model_devi_f_trust_lo":0.30,
+  "model_devi_f_trust_hi":100.00, 
+  "cvasp": false
+}

tests/test_check_examples.py

@@ -28,6 +28,7 @@
 input_files = (
     (init_reaction_jdata, p_examples / "init" / "reaction.json"),
     (simplify_jdata, p_examples / "simplify" / "qm7.json"),
+    (simplify_jdata, p_examples / "simplify-MAPbI3-scan-lebesgue" / "simplify_example" / "simplify.json"),
     (run_jdata, p_examples / "run" / "dp2.x-lammps-vasp" / "param_CH4_deepmd-kit-2.0.1.json"),
     #(run_jdata, p_examples / "run" / "dp2.x-gromacs-gaussian" / "param.json"),
     (run_jdata, p_examples / "run" / "dp1.x-lammps-vasp" / "CH4" / "param_CH4_deepmd-kit-1.1.0.json"),