Feature/sof 2002

example/job/input.json

@@ -0,0 +1,3 @@
+{
+    "...": "include(theory/ml/input.json)"
+}

example/job/model/method/workflow/unit/execution.json

@@ -5,8 +5,7 @@
     "compute": null,
     "input": [
         {
-            "content": "K_POINTS automatic\n2 2 2 0 0 0",
-            "name": "pw_scf.in"
+            "...": "include(execution/input.json)"
         }
     ]
 }

example/job/model/method/workflow/unit/execution/input.json

@@ -0,0 +1,4 @@
+{
+    "content": "K_POINTS automatic\n2 2 2 0 0 0",
+    "name": "pw_scf.in"
+}

example/job/theory/ml/input.json

@@ -0,0 +1,12 @@
+{
+    "method":{
+        "type": "linear",
+        "subtype": "least_squares"
+    },
+    "fitParameters": [
+        "sample_weight"
+    ],
+    "targetProperties":[
+        "band_gaps"
+    ]
+}

example/job/theory/ml/linear/input_data

@@ -0,0 +1,12 @@
+{
+  "c8zG3XcnnTybfP3QQ": {
+    "Ge": 0.3125,
+    "Si": 0.6875,
+    "lattice_volume": 327.1127749111
+  },
+  "wWK2XCyivQrG8j34Z": {
+    "Ge": 0.1875,
+    "Si": 0.8125,
+    "lattice_volume": 327.1127749111
+  }
+}

example/job/theory/ml/linear/parameters.json

@@ -0,0 +1,12 @@
+{
+    "property": "band_gaps",
+    "coefficients":[
+        -3.339e-01,
+        2.555e-01,
+        -1.005e-02
+    ],
+    "intercept": 0.45,
+    "modelData" : {
+            "...": "include(input_data)"
+    }
+}

example/job/theory/ml/linear/precision.json

@@ -0,0 +1,4 @@
+{
+    "trainingError": 0.1,
+    "cvError": 0.3
+}

schema/job/app.json

@@ -30,11 +30,13 @@
         },
         {
             "$ref": "file:software/vasp.json"
+        },
+        {
+            "$ref": "file:software/ml.json"
         }
     ],
     "required": [
         "name",
-        "flavor",
         "version",
         "exec"
     ]

schema/job/input.json

@@ -0,0 +1,10 @@
+{
+    "$schema": "http://json-schema.org/draft-04/schema#",
+    "title": "application schema",
+    "type": "object",
+    "oneOf": [
+        {
+            "$ref": "file:theory/ml/input.json"
+        }
+    ]
+}

schema/job/model.json

@@ -27,6 +27,9 @@
     "oneOf": [
         {
             "$ref": "file:theory/dft.json"
+        },
+        {
+            "$ref": "file:theory/ml.json"
         }
     ],
     "required": [

schema/job/model/method.json

@@ -24,10 +24,10 @@
     },
     "oneOf": [
         {
-            "$ref": "file:../theory/dft/pseudopotential.json"
+            "$ref": "file:../theory/dft/_methods.json"
         },
         {
-            "$ref": "file:../theory/dft/local-orbital.json"
+            "$ref": "file:../theory/ml/_methods.json"
         }
     ],
     "required": [

schema/job/model/method/workflow/unit/execution.json

@@ -15,20 +15,15 @@
             "description": "Unit's input.",
             "type": "array",
             "items": {
-                "type": "object",
-                "properties": {
-                    "name": {
-                        "description": "Input file name. e.g. pw_scf.in",
-                        "type": "string"
+                "oneOf": [
+                    {
+                        "description": "textual input (eg. for simulation engines)",
+                        "$ref": "file:execution/input.json"
                     },
-                    "content": {
-                        "description": "Content of the input file. e.g. &CONTROL    calculation='scf' ...",
-                        "type": "string"
+                    {
+                        "description": "object input (eg. for machine learning)",
+                        "$ref": "file:../../../../input.json"
                     }
-                },
-                "required": [
-                    "name",
-                    "content"
                 ]
             }
         }

schema/job/model/method/workflow/unit/execution/input.json

@@ -0,0 +1,19 @@
+{
+    "$schema": "http://json-schema.org/draft-04/schema#",
+    "title": "execution unit text input schema",
+    "type": "object",
+    "properties": {
+            "name": {
+                "description": "Input file name. e.g. pw_scf.in",
+                "type": "string"
+            },
+            "content": {
+                "description": "Content of the input file. e.g. &CONTROL    calculation='scf' ...",
+                "type": "string"
+            }
+    },
+    "required": [
+        "name",
+        "content"
+    ]
+}

schema/job/software/ml.json

@@ -0,0 +1,27 @@
+{
+    "$schema": "http://json-schema.org/draft-04/schema#",
+    "title": "default machine learning app schema",
+    "type": "object",
+    "properties": {
+        "name": {
+            "enum": [
+                "ml"
+            ]
+        },
+        "summary": {
+            "enum": [
+                "Default Machine Learning Engine"
+            ]
+        },
+        "version": {
+            "enum": [
+                "0.1.1"
+            ]
+        },
+        "exec": {
+            "enum": [
+                "rupy"
+            ]
+        }
+    }
+}

schema/job/theory/dft/_methods.json

@@ -0,0 +1,12 @@
+{
+    "$schema": "http://json-schema.org/draft-04/schema#",
+    "title": "List of methods",
+    "oneOf": [
+        {
+            "$ref": "file:local-orbital.json"
+        },
+        {
+            "$ref": "file:pseudopotential.json"
+        }
+    ]
+}

schema/job/theory/ml.json

@@ -0,0 +1,18 @@
+{
+    "$schema": "http://json-schema.org/draft-04/schema#",
+    "title": "machine learning schema",
+    "description": "machine learning schema",
+    "properties": {
+        "type": {
+            "enum": [
+                "machine_learning"
+            ]
+        },
+        "subtype": {
+            "enum": [
+                "supervised",
+                "unsupervised"
+            ]
+        }
+    }
+}

schema/job/theory/ml/_methods.json

@@ -0,0 +1,9 @@
+{
+    "$schema": "http://json-schema.org/draft-04/schema#",
+    "title": "List of methods",
+    "oneOf": [
+        {
+            "$ref": "file:linear.json"
+        }
+    ]
+}

schema/job/theory/ml/input.json

@@ -0,0 +1,36 @@
+{
+    "$schema": "http://json-schema.org/draft-04/schema#",
+    "title": "Execution unit input schema for machine learning",
+    "type": "object",
+    "properties": {
+        "method": {
+            "description": "method information",
+            "type": "object",
+            "properties": {
+                "type": {
+                    "description": "method type",
+                    "type": "string"
+                },
+                "subtype": {
+                    "description": "method subtype",
+                    "type": "string"
+                }
+            },
+            "required": [
+                "type",
+                "subtype"
+            ]
+        },
+        "fitParameters": {
+            "description": "additional parameters used for model fitting using method subtype",
+            "type": "array"
+        },
+        "targetProperties": {
+            "description": "target properties to predict",
+            "type": "array"
+        }
+    },
+    "required": [
+        "targetProperties"
+    ]
+}

schema/job/theory/ml/linear.json

@@ -0,0 +1,17 @@
+{
+    "$schema": "http://json-schema.org/draft-04/schema#",
+    "title": "linear methods schema",
+    "properties": {
+        "type": {
+            "enum": [
+                "least_squares"
+            ]
+        },
+        "precision": {
+            "$ref": "file:linear/precision.json"
+        },
+        "data":{
+            "$ref":"file:linear/parameters.json"
+        }
+    }
+}

schema/job/theory/ml/linear/inputData.json

@@ -0,0 +1,6 @@
+{
+    "$schema": "http://json-schema.org/draft-04/schema#",
+    "descritpion": "Input data used for ML method",
+    "title": "input data schema",
+    "type": "object"
+}

schema/job/theory/ml/linear/parameters.json

@@ -0,0 +1,26 @@
+{
+    "$schema": "http://json-schema.org/draft-04/schema#",
+    "title": "regression parameters schema",
+    "properties": {
+        "property": {
+            "type": "string",
+            "description": "name of the predicted or optimized characteristic"
+        },
+        "coefficients": {
+            "type": "array",
+            "description": "weights/coefficients of the function from linear or non-linear fit"
+        },
+        "intercept": {
+            "type": "number",
+            "description": "intercept (shift) from the linear fit of data points"
+        },
+        "inputData": {
+            "$ref":"file:inputData.json"
+        }
+    },
+    "required": [
+        "property",
+        "coefficients",
+        "intercept"
+    ]
+}

schema/job/theory/ml/linear/precision.json

@@ -0,0 +1,18 @@
+{
+    "$schema": "http://json-schema.org/draft-04/schema#",
+    "title": "precision schema for regression",
+    "properties": {
+        "trainingError": {
+            "description": "training error of the estimator",
+            "type": "number"
+        },
+        "cvError": {
+            "description": "cross validation error of the estimator",
+            "type": "number"
+        }
+    },
+    "required": [
+        "trainingError",
+        "cvError"
+    ]
+}