Commit df89ebb1 authored by fan jicong's avatar fan jicong Committed by Sujen

Master

parent 34db6d47
{
"problem": "4550_MiceProtein_problem",
"problem": "38_sick_problem",
"full_inputs": [
"4550_MiceProtein_dataset"
"38_sick_dataset"
],
"train_inputs": [
"4550_MiceProtein_dataset_TRAIN"
"38_sick_dataset_TRAIN"
],
"test_inputs": [
"4550_MiceProtein_dataset_TEST"
"38_sick_dataset_TEST"
],
"score_inputs": [
"4550_MiceProtein_dataset_SCORE"
"38_sick_dataset_SCORE"
]
}
{
"created": "2019-06-15T00:18:59.526485Z",
"digest": "62e7e6f214e4eb9ed4edf534e0d9b0c13f9426cdec30bedfbb8bc51db410265e",
"id": "2c80ae13-732d-43a4-ba06-1b064e644524",
"created": "2019-06-16T21:06:59.065355Z",
"digest": "6e0005297cc824ec76e96da63db7dc6f2f9223626da974e84e80b7f2ab6f1f0c",
"id": "6d7ee300-94e1-4a1c-94ca-a7e6b178b0f3",
"inputs": [
{
"name": "inputs"
......@@ -276,7 +276,7 @@
}
],
"primitive": {
"digest": "13c8e81e295b0ed8fa648c773c689a40246cbec9e7ee290fd0c1decb0842a562",
"digest": "a2d6e8307f6c98e89f7417a6401966d66c180bfdfd21c203e1e3ecee0a8f44f3",
"id": "e6ee30fa-af68-4bfe-9234-5ca7e7ac8e93",
"name": "Matrix Completion via Sparse Factorization",
"python_path": "d3m.primitives.collaborative_filtering.high_rank_imputer.Cornell",
......
{
"created": "2019-06-15T00:01:30.590331Z",
"digest": "23fdb1f2468ea47080ffeae2c7a310a6de36143031c190fee13b2967b9789014",
"id": "0be9e78f-b3d6-4f26-be5f-e6ec3ddb1027",
"created": "2019-06-16T21:03:04.045796Z",
"digest": "f4763762b6473ab1c4d4026cc5b8155f898553015d11f21a5d3edd6bd958fb77",
"id": "ae32ba4c-8103-4aff-b55c-bf10ed98639f",
"inputs": [
{
"name": "inputs"
......@@ -150,7 +150,7 @@
}
],
"primitive": {
"digest": "13c8e81e295b0ed8fa648c773c689a40246cbec9e7ee290fd0c1decb0842a562",
"digest": "a2d6e8307f6c98e89f7417a6401966d66c180bfdfd21c203e1e3ecee0a8f44f3",
"id": "e6ee30fa-af68-4bfe-9234-5ca7e7ac8e93",
"name": "Matrix Completion via Sparse Factorization",
"python_path": "d3m.primitives.collaborative_filtering.high_rank_imputer.Cornell",
......
{
"created": "2019-06-15T02:49:58.561288Z",
"digest": "a3749d5d33fd81e5892c9f80a97b558700b90867e6b10438cb8705626e61fa88",
"id": "de4354fb-8aa8-4176-b4bf-d58fde78b889",
"created": "2019-06-16T21:09:21.096784Z",
"digest": "b72b5f7d6a6585a9e10d2515867ce8c13c5ddfd224f5ff7c21c410818581a6a7",
"id": "be68ba08-5015-4699-888f-fe714242dea2",
"inputs": [
{
"name": "inputs"
......@@ -151,7 +151,30 @@
74,
75,
76,
77
77,
78,
79,
80,
81,
82,
83,
84,
85,
86,
87,
88,
89,
90,
91,
92,
93,
94,
95,
96,
97,
98,
99,
100
],
"type": "VALUE"
}
......@@ -204,14 +227,57 @@
}
],
"primitive": {
"digest": "9878fdeb255c5b4fb2beaf053e68b2913e3d7b1c26e40c530c1cb4fe562fde26",
"id": "d016df89-de62-3c53-87ed-c06bb6a23cde",
"name": "sklearn.impute.SimpleImputer",
"python_path": "d3m.primitives.data_cleaning.imputer.SKlearn",
"digest": "4183552e52f20262bfbb3711300a39216d15749d204cc7b475cad55f8dda1cfd",
"id": "d639947e-ece0-3a39-a666-e974acf4521d",
"name": "sklearn.preprocessing.data.StandardScaler",
"python_path": "d3m.primitives.data_preprocessing.standard_scaler.SKlearn",
"version": "2019.4.4"
},
"type": "PRIMITIVE"
},
{
"arguments": {
"inputs": {
"data": "steps.4.produce",
"type": "CONTAINER"
},
"outputs": {
"data": "steps.4.produce",
"type": "CONTAINER"
}
},
"hyperparams": {
"alpha": {
"data": 0.1,
"type": "VALUE"
},
"beta": {
"data": 0.1,
"type": "VALUE"
},
"d": {
"data": 50,
"type": "VALUE"
},
"maxiter": {
"data": 1000,
"type": "VALUE"
}
},
"outputs": [
{
"id": "produce"
}
],
"primitive": {
"digest": "a2d6e8307f6c98e89f7417a6401966d66c180bfdfd21c203e1e3ecee0a8f44f3",
"id": "e6ee30fa-af68-4bfe-9234-5ca7e7ac8e93",
"name": "Matrix Completion via Sparse Factorization",
"python_path": "d3m.primitives.collaborative_filtering.high_rank_imputer.Cornell",
"version": "v0.1.1"
},
"type": "PRIMITIVE"
},
{
"arguments": {
"inputs": {
......@@ -244,7 +310,7 @@
{
"arguments": {
"inputs": {
"data": "steps.5.produce",
"data": "steps.6.produce",
"type": "CONTAINER"
}
},
......@@ -265,39 +331,22 @@
{
"arguments": {
"inputs": {
"data": "steps.4.produce",
"data": "steps.5.produce",
"type": "CONTAINER"
},
"outputs": {
"data": "steps.7.produce",
"type": "CONTAINER"
}
},
"hyperparams": {
"k": {
"data": 2,
"C": {
"data": 1000,
"type": "VALUE"
}
},
"outputs": [
{
"id": "produce"
}
],
"primitive": {
"digest": "f1ac0dc92e4fb6405e5a2ba1c7771de9280a64daa1a24c14ee57f86cfb79c305",
"id": "7c357e6e-7124-4f2a-8371-8021c8c95cc9",
"name": "Huber PCA",
"python_path": "d3m.primitives.feature_extraction.huber_pca.Cornell",
"version": "v0.1.1"
},
"type": "PRIMITIVE"
},
{
"arguments": {
"inputs": {
"data": "steps.7.produce",
"type": "CONTAINER"
},
"outputs": {
"data": "steps.6.produce",
"type": "CONTAINER"
"kernel": {
"data": "rbf",
"type": "VALUE"
}
},
"outputs": [
......@@ -306,10 +355,10 @@
}
],
"primitive": {
"digest": "1e95597335ea675f941f08c916e586a414c0405a2ea0e3da0a0e3b1ee47ba761",
"id": "1dd82833-5692-39cb-84fb-2455683075f3",
"name": "sklearn.ensemble.forest.RandomForestClassifier",
"python_path": "d3m.primitives.classification.random_forest.SKlearn",
"digest": "7eedc5772884caaed449badc5b20095633f199fbc2e68c4939fa2f81eeaaf2b3",
"id": "0ae7d42d-f765-3348-a28c-57d94880aa6a",
"name": "sklearn.svm.classes.SVC",
"python_path": "d3m.primitives.classification.svc.SKlearn",
"version": "2019.4.4"
},
"type": "PRIMITIVE"
......@@ -323,7 +372,7 @@
},
"hyperparams": {
"encoder": {
"data": 6,
"data": 7,
"type": "PRIMITIVE"
}
},
......
{
"problem": "LL1_VTXC_1369_synthetic_problem",
"full_inputs": [
"LL1_VTXC_1369_synthetic_dataset"
],
"train_inputs": [
"LL1_VTXC_1369_synthetic_dataset_TRAIN"
],
"test_inputs": [
"LL1_VTXC_1369_synthetic_dataset_TEST"
],
"score_inputs": [
"LL1_VTXC_1369_synthetic_dataset_SCORE"
]
}
......@@ -18,7 +18,7 @@
"installation": [
{
"type": "PIP",
"package_uri": "git+https://github.com/cyangcornell/[email protected]87c6d370b902acb5e0326aeccccf35259bddb637#egg=pyglrm-d3m"
"package_uri": "git+https://github.com/cyangcornell/[email protected]a8f383d27f7dc878a92de0aa2337d869ff2d01b4#egg=pyglrm-d3m"
}
],
"python_path": "d3m.primitives.collaborative_filtering.high_rank_imputer.Cornell",
......@@ -230,10 +230,11 @@
"temporary_directory": "typing.Union[NoneType, str]"
},
"params": {
"X": "d3m.container.numpy.ndarray"
"X": "d3m.container.numpy.ndarray",
"A": "d3m.container.numpy.ndarray"
}
},
"structural_type": "pyglrm_d3m.high_rank_imputer.HighRankImputer",
"description": "This primitive imputes a dataset in which data points are drawn from multiple subspaces, which in pratice means the data have mutiple groups/classes. In such cases, the data matrices are often of high-rank. In such cases, Sparse Factorization based Matrix Completion (SFMC) can outperform classical low-rank matrix completion methods.\nThe optimization is solved via accelerated proximal alternating minimization (APALM). The NaNs in the input matrix will be regarded as missing entries. The algorithm will recover the missing entries and return the recovered matrix as output.\nThe method can be used for collaborative filtering (recommendation system) and data preprocessing.\n\nAttributes\n----------\nmetadata : PrimitiveMetadata\n Primitive's metadata. Available as a class attribute.\nlogger : Logger\n Primitive's logger. Available as a class attribute.\nhyperparams : Hyperparams\n Hyperparams passed to the constructor.\nrandom_seed : int\n Random seed passed to the constructor.\ndocker_containers : Dict[str, DockerContainer]\n A dict mapping Docker image keys from primitive's metadata to (named) tuples containing\n container's address under which the container is accessible by the primitive, and a\n dict mapping exposed ports to ports on that address.\nvolumes : Dict[str, str]\n A dict mapping volume keys from primitive's metadata to file and directory paths\n where downloaded and extracted files are available to the primitive.\ntemporary_directory : str\n An absolute path to a temporary directory a primitive can use to store any files\n for the duration of the current pipeline run phase. Directory is automatically\n cleaned up after the current pipeline run phase finishes.",
"digest": "13c8e81e295b0ed8fa648c773c689a40246cbec9e7ee290fd0c1decb0842a562"
"digest": "a2d6e8307f6c98e89f7417a6401966d66c180bfdfd21c203e1e3ecee0a8f44f3"
}
{
"created": "2019-06-15T00:33:15.123038Z",
"digest": "e3c31b9b297ccf1211bbaae93b1e0a8242070e24a8d75d70e182dc5b24a4ce81",
"id": "ebc772c6-1391-4bd3-9d61-68cfdb14e059",
"created": "2019-06-16T23:23:25.076377Z",
"digest": "2bb285ee805b41f5b6507b2905e343bc662388d7c7acb0493a1899a64b8e61db",
"id": "45ccc1a5-d135-4cc2-92b5-b4f6da75aade",
"inputs": [
{
"name": "inputs"
......@@ -250,7 +250,7 @@
},
"hyperparams": {
"k": {
"data": 30,
"data": 40,
"type": "VALUE"
}
},
......@@ -260,7 +260,7 @@
}
],
"primitive": {
"digest": "5cdadc933bbc2ba20603ea955b533418003f6eeabb42708ef739559d84319cc3",
"digest": "9c4e254ba8e26550ba61e0f3ce2da9516ed4e024a1563feef1e1367ca999d85b",
"id": "c959da5a-aa2e-44a6-86f2-a52fe2ab9db7",
"name": "Low Rank Imputer",
"python_path": "d3m.primitives.data_preprocessing.low_rank_imputer.Cornell",
......
......@@ -17,7 +17,7 @@
"installation": [
{
"type": "PIP",
"package_uri": "git+https://github.com/cyangcornell/[email protected]87c6d370b902acb5e0326aeccccf35259bddb637#egg=pyglrm-d3m"
"package_uri": "git+https://github.com/cyangcornell/[email protected]a8f383d27f7dc878a92de0aa2337d869ff2d01b4#egg=pyglrm-d3m"
}
],
"python_path": "d3m.primitives.data_preprocessing.low_rank_imputer.Cornell",
......@@ -194,5 +194,5 @@
},
"structural_type": "pyglrm_d3m.low_rank_imputer.LowRankImputer",
"description": "This primitive performs low rank imputation: rather than just imputing missing entries with, for example, means or medians of each feature, it recover missing entries based on low rank structure of the dataset.\n\nAttributes\n----------\nmetadata : PrimitiveMetadata\n Primitive's metadata. Available as a class attribute.\nlogger : Logger\n Primitive's logger. Available as a class attribute.\nhyperparams : Hyperparams\n Hyperparams passed to the constructor.\nrandom_seed : int\n Random seed passed to the constructor.\ndocker_containers : Dict[str, DockerContainer]\n A dict mapping Docker image keys from primitive's metadata to (named) tuples containing\n container's address under which the container is accessible by the primitive, and a\n dict mapping exposed ports to ports on that address.\nvolumes : Dict[str, str]\n A dict mapping volume keys from primitive's metadata to file and directory paths\n where downloaded and extracted files are available to the primitive.\ntemporary_directory : str\n An absolute path to a temporary directory a primitive can use to store any files\n for the duration of the current pipeline run phase. Directory is automatically\n cleaned up after the current pipeline run phase finishes.",
"digest": "5cdadc933bbc2ba20603ea955b533418003f6eeabb42708ef739559d84319cc3"
"digest": "c9a2cf9bb86cc149f9ab895a1be5ddae845dac09068f291f54a54c422eb9fc56"
}
......@@ -16,7 +16,7 @@
"installation": [
{
"type": "PIP",
"package_uri": "git+https://github.com/cyangcornell/[email protected]87c6d370b902acb5e0326aeccccf35259bddb637#egg=pyglrm-d3m"
"package_uri": "git+https://github.com/cyangcornell/[email protected]955890f2e42154adb3584a4b0bb3bf1a9c913d84#egg=pyglrm-d3m"
}
],
"python_path": "d3m.primitives.feature_extraction.huber_pca.Cornell",
......@@ -51,6 +51,51 @@
"https://metadata.datadrivendiscovery.org/types/TuningParameter"
],
"description": "Maximum rank of the decomposed matrices. For example, if the matrix A to be decomposed is m-by-n, then after decomposition A\u2248XY, X is m-by-k, Y is k-by-n. "
},
"huber_crossover": {
"type": "d3m.metadata.hyperparams.Hyperparameter",
"default": 1.0,
"structural_type": "float",
"semantic_types": [
"https://metadata.datadrivendiscovery.org/types/TuningParameter"
],
"description": "Crossover for the Huber loss."
},
"rx": {
"type": "d3m.metadata.hyperparams.Hyperparameter",
"default": "ZeroReg",
"structural_type": "str",
"semantic_types": [
"https://metadata.datadrivendiscovery.org/types/TuningParameter"
],
"description": "Regularizer for X."
},
"ry": {
"type": "d3m.metadata.hyperparams.Hyperparameter",
"default": "ZeroReg",
"structural_type": "str",
"semantic_types": [
"https://metadata.datadrivendiscovery.org/types/TuningParameter"
],
"description": "Regularizer for Y."
},
"lambda_x": {
"type": "d3m.metadata.hyperparams.Hyperparameter",
"default": 1.0,
"structural_type": "float",
"semantic_types": [
"https://metadata.datadrivendiscovery.org/types/TuningParameter"
],
"description": "Coefficient of rx."
},
"lambda_y": {
"type": "d3m.metadata.hyperparams.Hyperparameter",
"default": 1.0,
"structural_type": "float",
"semantic_types": [
"https://metadata.datadrivendiscovery.org/types/TuningParameter"
],
"description": "Coefficient of ry."
}
},
"arguments": {
......@@ -201,5 +246,5 @@
},
"structural_type": "pyglrm_d3m.huber_pca.HuberPCA",
"description": "By Huber PCA, this primitive gets low rank representation of the original dataset via Huber loss (rather than the L2 loss in standard PCA), and is thus more robust to outliers.\n\nAttributes\n----------\nmetadata : PrimitiveMetadata\n Primitive's metadata. Available as a class attribute.\nlogger : Logger\n Primitive's logger. Available as a class attribute.\nhyperparams : Hyperparams\n Hyperparams passed to the constructor.\nrandom_seed : int\n Random seed passed to the constructor.\ndocker_containers : Dict[str, DockerContainer]\n A dict mapping Docker image keys from primitive's metadata to (named) tuples containing\n container's address under which the container is accessible by the primitive, and a\n dict mapping exposed ports to ports on that address.\nvolumes : Dict[str, str]\n A dict mapping volume keys from primitive's metadata to file and directory paths\n where downloaded and extracted files are available to the primitive.\ntemporary_directory : str\n An absolute path to a temporary directory a primitive can use to store any files\n for the duration of the current pipeline run phase. Directory is automatically\n cleaned up after the current pipeline run phase finishes.",
"digest": "f1ac0dc92e4fb6405e5a2ba1c7771de9280a64daa1a24c14ee57f86cfb79c305"
"digest": "473977bc8642f374c203f18c2169e71697395e82354186f41931be5ffc766561"
}
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment