Skip to content

Commit 817710b

Browse files
md-shafiul-alammd-shafiul-alam
committed
bf16
1 parent e3d9a35 commit 817710b

File tree

9 files changed

+266
-2
lines changed

9 files changed

+266
-2
lines changed

‎configs/regular/bf16/dbscan.json

Lines changed: 41 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,41 @@
1+
{
2+
"INCLUDE": ["../common/sklearn.json"],
3+
"PARAMETERS_SETS": {
4+
"common dbscan parameters": {
5+
"algorithm": {
6+
"estimator": "DBSCAN",
7+
"estimator_params": {
8+
"eps": "[SPECIAL_VALUE]distances_quantile:0.01",
9+
"min_samples": 5,
10+
"metric": "euclidean"
11+
}
12+
},
13+
"data": {
14+
"dtype": ["float32"]
15+
}
16+
},
17+
"sklearn dbscan parameters": {
18+
"algorithm": {
19+
"estimator_params": {
20+
"algorithm": "brute",
21+
"n_jobs": "[SPECIAL_VALUE]physical_cpus"
22+
}
23+
}
24+
},
25+
"synthetic dataset": {
26+
"data": [
27+
{ "source": "make_blobs", "generation_kwargs": { "n_samples": 100000, "n_features": 10, "centers": 10 }, "algorithm": { "eps": 5, "min_samples": 5 } }
28+
]
29+
}
30+
},
31+
"TEMPLATES": {
32+
"sklearn dbscan": {
33+
"SETS": [
34+
"sklearn-ex[gpu] implementations",
35+
"common dbscan parameters",
36+
"sklearn dbscan parameters",
37+
"synthetic dataset"
38+
]
39+
}
40+
}
41+
}

‎configs/regular/bf16/ensemble.json

Lines changed: 45 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,45 @@
1+
{
2+
"INCLUDE": ["../common/sklearn.json"],
3+
"PARAMETERS_SETS": {
4+
"common ensemble params": {
5+
"algorithm": {
6+
"estimator_params": {
7+
"n_estimators": 200,
8+
"max_depth": 16,
9+
"max_samples": 1.0,
10+
"min_samples_split": 5,
11+
"min_samples_leaf": 2,
12+
"min_impurity_decrease": 0.0,
13+
"bootstrap": true,
14+
"random_state": 42
15+
}
16+
}
17+
},
18+
"sklearn ensemble classifier params": {
19+
"algorithm": {
20+
"estimator": ["RandomForestClassifier", "ExtraTreesClassifier"],
21+
"estimator_params": {
22+
"criterion": "gini",
23+
"max_features": "sqrt",
24+
"max_leaf_nodes": null,
25+
"n_jobs": "[SPECIAL_VALUE]physical_cpus"
26+
}
27+
}
28+
},
29+
"synthetic data": {
30+
"data": [
31+
{ "source": "make_classification", "split_kwargs": { "train_size": 500000, "test_size": 1000 }, "generation_kwargs": { "n_samples": 501000, "n_features": 10, "n_classes": 2 }, "algorithm": { "estimator_params": { "n_estimators": 20, "max_depth": 4 } } }
32+
]
33+
}
34+
},
35+
"TEMPLATES": {
36+
"sklearn ensemble classification": {
37+
"SETS": [
38+
"sklearn-ex[gpu] implementations",
39+
"common ensemble params",
40+
"sklearn ensemble classifier params",
41+
"synthetic data"
42+
]
43+
}
44+
}
45+
}

‎configs/regular/bf16/kmeans.json

Lines changed: 40 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,40 @@
1+
{
2+
"INCLUDE": ["../common/sklearn.json"],
3+
"PARAMETERS_SETS": {
4+
"common kmeans parameters": {
5+
"algorithm": {
6+
"estimator": "KMeans",
7+
"estimator_params": {
8+
"n_clusters": "[SPECIAL_VALUE]auto",
9+
"n_init": 1,
10+
"max_iter": 30,
11+
"tol": 1e-3,
12+
"random_state": 42
13+
},
14+
"estimator_methods": { "inference": "predict" }
15+
},
16+
"data": {
17+
"dtype": ["float32", "float64"],
18+
"preprocessing_kwargs": { "normalize": true }
19+
}
20+
},
21+
"sklearn kmeans parameters": {
22+
"algorithm": { "estimator_params": { "init": "k-means++", "algorithm": "lloyd" } }
23+
},
24+
"synthetic data": {
25+
"data": [
26+
{ "source": "make_blobs", "generation_kwargs": { "n_samples": 1000000, "n_features": 100, "centers": 100 }, "algorithm": { "n_clusters": 100, "max_iter": 100 } }
27+
]
28+
}
29+
},
30+
"TEMPLATES": {
31+
"sklearn kmeans": {
32+
"SETS": [
33+
"sklearn-ex[gpu] implementations",
34+
"common kmeans parameters",
35+
"sklearn kmeans parameters",
36+
"synthetic data"
37+
]
38+
}
39+
}
40+
}

‎configs/regular/bf16/knn.json

Lines changed: 34 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,34 @@
1+
{
2+
"INCLUDE": ["../common/sklearn.json"],
3+
"PARAMETERS_SETS": {
4+
"common knn parameters": {
5+
"algorithm": {
6+
"estimator_params": {
7+
"n_neighbors": [10, 100],
8+
"weights": "uniform"
9+
}
10+
},
11+
"data": {
12+
"preprocessing_kwargs": { "normalize": true }
13+
}
14+
},
15+
"sklearn knn parameters": {
16+
"algorithm": { "estimator_params": { "n_jobs": "[SPECIAL_VALUE]physical_cpus" } }
17+
},
18+
"synthetic classification data": {
19+
"data": [
20+
{ "source": "make_classification", "split_kwargs": { "train_size": 500000, "test_size": 50000 }, "generation_kwargs": { "n_samples": 550000, "n_features": 100, "n_classes": 2, "n_informative": "[SPECIAL_VALUE]0.5" } }
21+
]
22+
}
23+
},
24+
"TEMPLATES": {
25+
"sklearn brute knn clsf": {
26+
"SETS": [
27+
"sklearn-ex[gpu] implementations",
28+
"common knn parameters",
29+
"sklearn knn parameters",
30+
"synthetic classification data"
31+
]
32+
}
33+
}
34+
}

‎configs/regular/bf16/linear_model.json

Lines changed: 29 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,29 @@
1+
{
2+
"INCLUDE": ["../common/sklearn.json"],
3+
"PARAMETERS_SETS": {
4+
"synthetic data": {
5+
"data": [
6+
{ "source": "make_regression", "generation_kwargs": { "n_samples": 3005000, "n_features": 10, "noise": 1.25 }, "split_kwargs": { "train_size": 3000000, "test_size": 5000 } }
7+
]
8+
},
9+
"common linear parameters": {
10+
"algorithm": {
11+
"estimator": "LinearRegression",
12+
"estimator_params": { "fit_intercept": true, "copy_X": true }
13+
}
14+
},
15+
"sklearn linear parameters": {
16+
"estimator_params": { "n_jobs": "[SPECIAL_VALUE]physical_cpus" }
17+
}
18+
},
19+
"TEMPLATES": {
20+
"sklearn linear": {
21+
"SETS": [
22+
"sklearn-ex[gpu] implementations",
23+
"common linear parameters",
24+
"sklearn linear parameters",
25+
"synthetic data"
26+
]
27+
}
28+
}
29+
}

‎configs/regular/bf16/logreg.json

Lines changed: 42 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,42 @@
1+
{
2+
"INCLUDE": ["../common/sklearn.json"],
3+
"PARAMETERS_SETS": {
4+
"common logreg parameters": {
5+
"algorithm": {
6+
"estimator": "LogisticRegression",
7+
"estimator_methods": { "inference": "predict" },
8+
"estimator_params": {
9+
"penalty": "l2",
10+
"tol": 1e-4,
11+
"C": 1.0,
12+
"l1_ratio": null,
13+
"max_iter": 200
14+
}
15+
}
16+
},
17+
"sklearn logreg parameters": {
18+
"algorithm": {
19+
"estimator_params": {
20+
"solver": "lbfgs",
21+
"n_jobs": "[SPECIAL_VALUE]physical_cpus",
22+
"random_state": 42
23+
}
24+
}
25+
},
26+
"synthetic data": {
27+
"data": [
28+
{ "source": "make_classification", "split_kwargs": { "train_size": 5000000, "test_size": 1000 }, "generation_kwargs": { "n_samples": 5001000, "n_features": 10, "n_classes": 2 } }
29+
]
30+
}
31+
},
32+
"TEMPLATES": {
33+
"sklearn logreg": {
34+
"SETS": [
35+
"sklearn-ex[gpu] implementations",
36+
"common logreg parameters",
37+
"sklearn logreg parameters",
38+
"synthetic data"
39+
]
40+
}
41+
}
42+
}

‎configs/regular/bf16/pca.json

Lines changed: 33 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,33 @@
1+
{
2+
"INCLUDE": ["../common/sklearn.json"],
3+
"PARAMETERS_SETS": {
4+
"pca parameters": {
5+
"algorithm": {
6+
"estimator": "PCA",
7+
"estimator_params": {
8+
"n_components": 3,
9+
"copy": true,
10+
"whiten": false,
11+
"svd_solver": "covariance_eigh",
12+
"tol": 0.0,
13+
"iterated_power": 15,
14+
"random_state": 42
15+
}
16+
}
17+
},
18+
"synthetic data": {
19+
"data": [
20+
{ "source": "make_blobs", "generation_kwargs": { "n_samples": 10000000, "n_features": 10, "centers": 1 } }
21+
]
22+
}
23+
},
24+
"TEMPLATES": {
25+
"sklearn pca": {
26+
"SETS": [
27+
"sklearn-ex[gpu] implementations",
28+
"pca parameters",
29+
"synthetic data"
30+
]
31+
}
32+
}
33+
}

‎configs/spmd/large_scale/linear_model.json

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -10,7 +10,7 @@
1010
"synthetic data": {
1111
"data": [
1212
{ "source": "make_regression", "generation_kwargs": { "n_samples": 30005000, "n_features": 10, "noise": 1.25 }, "split_kwargs": { "train_size": 30000000, "test_size": 5000 } },
13-
{ "source": "make_regression", "generation_kwargs": { "n_samples": 305000, "n_features": 1000, "noise": 1.25 }, "split_kwargs": { "train_size": 300000, "test_size": 5000 } }
13+
{ "source": "make_regression", "generation_kwargs": { "n_samples": 305000, "n_features": 1000, "noise": 1.25 }, "split_kwargs": { "train_size": 300000, "test_size": 5000 } }
1414
]
1515
}
1616
},

‎configs/spmd/large_scale/logreg.json

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -10,7 +10,7 @@
1010
},
1111
"synthetic data": {
1212
"data": [
13-
{ "source": "make_classification", "split_kwargs": { "train_size": 5000000, "test_size": 1000 }, "generation_kwargs": { "n_samples": 5001000, "n_features": 10, "n_classes": 2 } },
13+
{ "source": "make_classification", "split_kwargs": { "train_size": 5000000, "test_size": 1000 }, "generation_kwargs": { "n_samples": 5001000, "n_features": 10, "n_classes": 2 } },
1414
{ "source": "make_classification", "split_kwargs": { "train_size": 100000, "test_size": 1000 }, "generation_kwargs": { "n_samples": 101000, "n_features": 1000, "n_classes": 2 } }
1515
]
1616
}

0 commit comments

Comments
 (0)