@mat3ra/esse 2023.12.25-2

package/schema/software_directory/ml/unit/processing/data_transformation/manipulation.json

@@ -0,0 +1,50 @@
+{
+    "$id": "software-directory/ml/unit/processing/data-transformation/manipulation",
+    "$schema": "http://json-schema.org/draft-04/schema#",
+    "title": "manipulation unit schema",
+    "allOf": [
+        {
+            "$ref": "../../../../../workflow/unit/processing.json"
+        }
+    ],
+    "type": "object",
+    "properties": {
+        "operation": {
+            "type": "string",
+            "enum": [
+                "data_transformation"
+            ]
+        },
+        "operationType": {
+            "type": "string",
+            "enum": [
+                "manipulation"
+            ]
+        },
+        "inputData": {
+            "type": "object",
+            "properties": {
+                "cleanMissingData": {
+                    "description": "whether to clean missing data, eg. NaN",
+                    "type": "boolean",
+                    "default": true
+                },
+                "removeDuplicateRows": {
+                    "description": "whether to remove duplicate rows",
+                    "type": "boolean",
+                    "default": true
+                },
+                "replaceNoneValuesWith": {
+                    "description": "replace None values with a given value",
+                    "default": 0,
+                    "type": "number"
+                }
+            },
+            "required": [
+                "cleanMissingData",
+                "removeDuplicateRows",
+                "replaceNoneValuesWith"
+            ]
+        }
+    }
+}

package/schema/software_directory/ml/unit/processing/data_transformation/scale_and_reduce.json

@@ -0,0 +1,70 @@
+{
+    "$id": "software-directory/ml/unit/processing/data-transformation/scale-and-reduce",
+    "$schema": "http://json-schema.org/draft-04/schema#",
+    "title": "scale and reduce unit schema",
+    "allOf": [
+        {
+            "$ref": "../../../../../workflow/unit/processing.json"
+        }
+    ],
+    "type": "object",
+    "properties": {
+        "operation": {
+            "type": "string",
+            "enum": [
+                "data_transformation"
+            ]
+        },
+        "operationType": {
+            "type": "string",
+            "enum": [
+                "scale_and_reduce"
+            ]
+        },
+        "inputData": {
+            "type": "object",
+            "properties": {
+                "scaler": {
+                    "description": "type of scaler to be applied",
+                    "type": "string",
+                    "default": "standard_scaler",
+                    "enum": [
+                        "standard_scaler"
+                    ]
+                },
+                "perFeature": {
+                    "decription": "per-feature scaling data",
+                    "type": "array",
+                    "items": {
+                        "type": "object",
+                        "properties": {
+                            "variance": {
+                                "description": "variance in original training data",
+                                "type": "number"
+                            },
+                            "mean": {
+                                "description": "mean value of the original training data",
+                                "type": "number"
+                            },
+                            "scale": {
+                                "description": "scale multiplier for this feature/property",
+                                "type": "number"
+                            },
+                            "name": {
+                                "description": "feature/property name in 'flattened' format",
+                                "type": "string"
+                            }
+                        },
+                        "required": [
+                            "name",
+                            "scale"
+                        ]
+                    }
+                }
+            },
+            "required": [
+                "scaler"
+            ]
+        }
+    }
+}

package/schema/software_directory/ml/unit/processing/data_transformation.json

@@ -0,0 +1,8 @@
+{
+    "$id": "software-directory/ml/unit/processing/data-transformation",
+    "oneOf": [
+        {
+            "$ref": "data_transformation/scale_and_reduce.json"
+        }
+    ]
+}

package/schema/software_directory/ml/unit/processing/feature_selection/filter_based.json

@@ -0,0 +1,45 @@
+{
+    "$id": "software-directory/ml/unit/processing/feature-selection/filter-based",
+    "$schema": "http://json-schema.org/draft-04/schema#",
+    "title": "filter-based feature selection unit schema",
+    "allOf": [
+        {
+            "$ref": "../../../../../workflow/unit/processing.json"
+        }
+    ],
+    "type": "object",
+    "properties": {
+        "operation": {
+            "type": "string",
+            "enum": [
+                "feature_selection"
+            ]
+        },
+        "operationType": {
+            "type": "string",
+            "enum": [
+                "filter_based"
+            ]
+        },
+        "inputData": {
+            "type": "object",
+            "properties": {
+                "nFeatures": {
+                    "description": "number of features to select for model training. If equal to 0, will use all available features",
+                    "type": "number"
+                },
+                "algorithm": {
+                    "description": "feature selection algorithm following sklearn.feature_selection",
+                    "type": "string",
+                    "enum": [
+                        "f_regression"
+                    ]
+                }
+            },
+            "required": [
+                "nFeatures",
+                "algorithm"
+            ]
+        }
+    }
+}

package/schema/software_directory/ml/unit/processing/feature_selection.json

@@ -0,0 +1,8 @@
+{
+    "$id": "software-directory/ml/unit/processing/feature-selection",
+    "oneOf": [
+        {
+            "$ref": "feature_selection/filter_based.json"
+        }
+    ]
+}

package/schema/software_directory/ml/unit/processing.json

@@ -0,0 +1,11 @@
+{
+    "$id": "software-directory/ml/unit/processing",
+    "oneOf": [
+        {
+            "$ref": "processing/data_transformation.json"
+        },
+        {
+            "$ref": "processing/feature_selection.json"
+        }
+    ]
+}

package/schema/software_directory/modeling/espresso/arguments.json

@@ -0,0 +1,44 @@
+{
+    "$id": "software-directory/modeling/espresso/arguments",
+    "$schema": "http://json-schema.org/draft-04/schema#",
+    "title": "quantum espresso arguments schema",
+    "type": "object",
+    "properties": {
+        "nimage": {
+            "description": "Processors can be divided into different `images`, each corresponding to a different self-consistent or linear-response calculation, loosely coupled to others.",
+            "type": "integer",
+            "default": 1,
+            "minimum": 1,
+            "maximum": 100
+        },
+        "npools": {
+            "description": "Each image can be subpartitioned into `pools`, each taking care of a group of k-points.",
+            "type": "integer",
+            "default": 1,
+            "minimum": 1,
+            "maximum": 100
+        },
+        "nband": {
+            "description": "Each pool is subpartitioned into `band groups`, each taking care of a group of Kohn-Sham orbitals (also called bands, or wavefunctions).",
+            "type": "integer",
+            "default": 1,
+            "minimum": 1,
+            "maximum": 100
+        },
+        "ntg": {
+            "description": "In order to allow good parallelization of the 3D FFT when the number of processors exceeds the number of FFT planes, FFTs on Kohn-Sham states are redistributed to `task` groups so that each group can process several wavefunctions at the same time.",
+            "type": "integer",
+            "default": 1,
+            "minimum": 1,
+            "maximum": 100
+        },
+        "ndiag": {
+            "description": "A further level of parallelization, independent on PW or k-point parallelization, is the parallelization of subspace diagonalization / iterative orthonormalization. Both operations required the diagonalization of arrays whose dimension is the number of Kohn-Sham states (or a small multiple of it). All such arrays are distributed block-like across the `linear-algebra group`, a subgroup of the pool of processors, organized in a square 2D grid. As a consequence the number of processors in the linear-algebra group is given by n2, where n is an integer; n2 must be smaller than the number of processors in the PW group. The diagonalization is then performed in parallel using standard linear algebra operations.",
+            "type": "integer",
+            "default": 1,
+            "minimum": 1,
+            "maximum": 100
+        }
+    },
+    "additionalProperties": false
+}

package/schema/software_directory/modeling/espresso.json

@@ -0,0 +1,26 @@
+{
+    "$id": "software-directory/modeling/espresso",
+    "$schema": "http://json-schema.org/draft-04/schema#",
+    "title": "espresso app schema",
+    "type": "object",
+    "properties": {
+        "name": {
+            "enum": [
+                "espresso"
+            ]
+        },
+        "summary": {
+            "enum": [
+                "Quantum Espresso"
+            ]
+        },
+        "version": {
+            "enum": [
+                "5.1.1",
+                "5.2.1",
+                "5.4.0",
+                "6.0.0"
+            ]
+        }
+    }
+}

package/schema/software_directory/modeling/nwchem.json

@@ -0,0 +1,34 @@
+{
+    "$id": "software-directory/modeling/nwchem",
+    "$schema": "http://json-schema.org/draft-04/schema#",
+    "title": "NWChem",
+    "type": "object",
+    "allOf": [
+        {
+            "$ref": "../../software/application.json"
+        }
+    ],
+    "properties": {
+        "name": {
+            "enum": [
+                "NWChem"
+            ]
+        },
+        "summary": {
+            "enum": [
+                "NWChem: a comprehensive and scalable open-source solution for large scale molecular simulations"
+            ]
+        },
+        "version": {
+            "enum": [
+                "6.6",
+                "7.0.2"
+            ]
+        },
+        "exec": {
+            "enum": [
+                "nwchem"
+            ]
+        }
+    }
+}

package/schema/software_directory/modeling/unit/execution.json

@@ -0,0 +1,14 @@
+{
+    "$id": "software-directory/modeling/unit/execution",
+    "$schema": "http://json-schema.org/draft-04/schema#",
+    "title": "execution unit schema for physics-based simulation engines (defined using espresso as example)",
+    "type": "object",
+    "allOf": [
+        {
+            "$ref": "../../../workflow/unit/execution.json"
+        },
+        {
+            "$ref": "../../../workflow/unit/input/_input.json"
+        }
+    ]
+}

package/schema/software_directory/modeling/vasp.json

@@ -0,0 +1,40 @@
+{
+    "$id": "software-directory/modeling/vasp",
+    "$schema": "http://json-schema.org/draft-04/schema#",
+    "title": "vienna ab-inito simulation package",
+    "type": "object",
+    "allOf": [
+        {
+            "$ref": "../../software/application.json"
+        }
+    ],
+    "properties": {
+        "name": {
+            "enum": [
+                "vasp"
+            ]
+        },
+        "summary": {
+            "enum": [
+                "vienna ab-initio simulation package"
+            ]
+        },
+        "flavor": {
+            "enum": [
+                "vasp",
+                "vasp_nscf",
+                "vasp_bands"
+            ]
+        },
+        "version": {
+            "enum": [
+                "5.3.5"
+            ]
+        },
+        "exec": {
+            "enum": [
+                "vasp"
+            ]
+        }
+    }
+}

package/schema/software_directory/scripting/jupyter-lab.json

@@ -0,0 +1,38 @@
+{
+    "$id": "software-directory/scripting/jupyter-lab",
+    "$schema": "http://json-schema.org/draft-04/schema#",
+    "title": "Jupyter Lab Application Schema",
+    "type": "object",
+    "allOf": [
+        {
+            "$ref": "../../software/application.json"
+        }
+    ],
+    "properties": {
+        "name": {
+            "enum": [
+                "jupyterLab"
+            ]
+        },
+        "flavor": {
+            "enum": [
+                "notebook"
+            ]
+        },
+        "summary": {
+            "enum": [
+                "Jupyter Lab"
+            ]
+        },
+        "version": {
+            "enum": [
+                "0.33.12"
+            ]
+        },
+        "exec": {
+            "enum": [
+                "jupyter"
+            ]
+        }
+    }
+}

package/schema/software_directory/scripting/python.json

@@ -0,0 +1,52 @@
+{
+    "$id": "software-directory/scripting/python",
+    "$schema": "http://json-schema.org/draft-04/schema#",
+    "title": "Python Programing Language Schema",
+    "type": "object",
+    "allOf": [
+        {
+            "$ref": "../../software/application.json"
+        }
+    ],
+    "properties": {
+        "name": {
+            "enum": [
+                "python"
+            ]
+        },
+        "flavor": {
+            "enum": [
+                "python2",
+                "python3"
+            ]
+        },
+        "summary": {
+            "enum": [
+                "Python Script"
+            ]
+        },
+        "version": {
+            "enum": [
+                "2.7.5",
+                "3.6.1"
+            ]
+        },
+        "exec": {
+            "enum": [
+                "python"
+            ]
+        },
+        "arguments": {
+            "description": "Optional arguments passed to the Python script",
+            "type": "string"
+        },
+        "environment": {
+            "description": "Optional environment variables exported before running the Python script",
+            "type": "object"
+        },
+        "dependencies": {
+            "description": "Optional Python dependencies, e.g. amqp==1.4.6",
+            "type": "array"
+        }
+    }
+}

package/schema/software_directory/scripting/shell.json

@@ -0,0 +1,52 @@
+{
+    "$id": "software-directory/scripting/shell",
+    "$schema": "http://json-schema.org/draft-04/schema#",
+    "title": "Shell Scripting Language Schema",
+    "type": "object",
+    "allOf": [
+        {
+            "$ref": "../../software/application.json"
+        }
+    ],
+    "properties": {
+        "name": {
+            "enum": [
+                "shell"
+            ]
+        },
+        "flavor": {
+            "enum": [
+                "sh",
+                "bash",
+                "zsh",
+                "csh"
+            ]
+        },
+        "summary": {
+            "enum": [
+                "Shell Script"
+            ]
+        },
+        "version": {
+            "enum": [
+                "4.2.46"
+            ]
+        },
+        "exec": {
+            "enum": [
+                "sh",
+                "bash",
+                "zsh",
+                "csh"
+            ]
+        },
+        "arguments": {
+            "description": "Optional arguments passed to the Shell script",
+            "type": "string"
+        },
+        "environment": {
+            "description": "Optional environment variables exported before running the Shell script",
+            "type": "object"
+        }
+    }
+}

package/schema/software_directory/scripting/unit/execution.json

@@ -0,0 +1,14 @@
+{
+    "$id": "software-directory/scripting/unit/execution",
+    "$schema": "http://json-schema.org/draft-04/schema#",
+    "title": "execution unit schema for scripting-based applications",
+    "type": "object",
+    "allOf": [
+        {
+            "$ref": "../../../workflow/unit/execution.json"
+        },
+        {
+            "$ref": "../../../workflow/unit/input/_input.json"
+        }
+    ]
+}

package/schema/system/_material.json

@@ -0,0 +1,19 @@
+{
+    "$id": "system/-material",
+    "$schema": "http://json-schema.org/draft-04/schema#",
+    "title": "Material entity reference schema",
+    "allOf": [
+        {
+            "$ref": "entity_reference.json"
+        }
+    ],
+    "properties": {
+        "cls": {
+            "description": "Material class",
+            "type": "string",
+            "enum": [
+                "Material"
+            ]
+        }
+    }
+}

package/schema/system/_parent_job.json

@@ -0,0 +1,19 @@
+{
+    "$id": "system/-parent-job",
+    "$schema": "http://json-schema.org/draft-04/schema#",
+    "title": "Parent job entity reference schema",
+    "allOf": [
+        {
+            "$ref": "entity_reference.json"
+        }
+    ],
+    "properties": {
+        "cls": {
+            "description": "Job class",
+            "type": "string",
+            "enum": [
+                "Job"
+            ]
+        }
+    }
+}

package/schema/system/_project.json

@@ -0,0 +1,19 @@
+{
+    "$id": "system/-project",
+    "$schema": "http://json-schema.org/draft-04/schema#",
+    "title": "Project entity reference schema",
+    "allOf": [
+        {
+            "$ref": "entity_reference.json"
+        }
+    ],
+    "properties": {
+        "cls": {
+            "description": "Project class",
+            "type": "string",
+            "enum": [
+                "Project"
+            ]
+        }
+    }
+}

package/schema/system/bankable.json

@@ -0,0 +1,15 @@
+{
+    "$id": "system/bankable",
+    "$schema": "http://json-schema.org/draft-04/schema#",
+    "title": "bankable schema",
+    "properties": {
+        "exabyteId": {
+            "description": "Identity of the corresponding bank entity",
+            "type": "string"
+        },
+        "hash": {
+            "description": "Hash string which is calculated based on the meaningful fields of the entity. Used to identify equal entities.",
+            "type": "string"
+        }
+    }
+}

package/schema/system/consistency_check.json

@@ -0,0 +1,39 @@
+{
+    "$id": "system/consistency-check",
+    "$schema": "http://json-schema.org/draft-04/schema#",
+    "title": "consistency check",
+    "type": "object",
+    "description": "The output of consistency checks performed on data adhering to JSON schema, but inconsistent with scientific or logical rules, to show problems in UI.",
+    "properties": {
+        "key": {
+            "type": "string",
+            "description": "Key of the property of the entity on which the consistency check is performed in Mongo dot notation, e.g. 'basis.coordinates.1'"
+        },
+        "name": {
+            "enum": [
+                "default",
+                "atomsTooClose",
+                "atomsOverlap"
+            ],
+            "description": "Name of the consistency check that is performed, which is listed in an enum."
+        },
+        "severity": {
+            "enum": [
+                "info",
+                "warning",
+                "error"
+            ],
+            "description": "Severity level of the problem, which is used in UI to differentiate."
+        },
+        "message": {
+            "type": "string",
+            "description": "Message generated by the consistency check describing the problem."
+        }
+    },
+    "required": [
+        "key",
+        "name",
+        "severity",
+        "message"
+    ]
+}

package/schema/system/creator.json

@@ -0,0 +1,19 @@
+{
+    "$id": "system/creator",
+    "$schema": "http://json-schema.org/draft-04/schema#",
+    "title": "Creator entity reference schema",
+    "allOf": [
+        {
+            "$ref": "entity_reference.json"
+        }
+    ],
+    "properties": {
+        "cls": {
+            "description": "Creator class",
+            "type": "string",
+            "enum": [
+                "User"
+            ]
+        }
+    }
+}

package/schema/system/creator_account.json

@@ -0,0 +1,10 @@
+{
+    "$id": "system/creator-account",
+    "$schema": "http://json-schema.org/draft-04/schema#",
+    "title": "creator account schema",
+    "properties": {
+        "creatorAccount": {
+            "$ref": "entity_reference.json"
+        }
+    }
+}