rbms 0.3.2__tar.gz → 0.4.0__tar.gz
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- {rbms-0.3.2 → rbms-0.4.0}/PKG-INFO +1 -1
- {rbms-0.3.2 → rbms-0.4.0}/pyproject.toml +1 -1
- {rbms-0.3.2 → rbms-0.4.0}/rbms/bernoulli_bernoulli/classes.py +5 -1
- {rbms-0.3.2 → rbms-0.4.0}/rbms/bernoulli_bernoulli/functional.py +9 -1
- {rbms-0.3.2 → rbms-0.4.0}/rbms/bernoulli_bernoulli/implement.py +12 -0
- {rbms-0.3.2 → rbms-0.4.0}/rbms/classes.py +8 -1
- rbms-0.4.0/rbms/dataset/__init__.py +83 -0
- {rbms-0.3.2 → rbms-0.4.0}/rbms/io.py +2 -0
- {rbms-0.3.2 → rbms-0.4.0}/rbms/parser.py +26 -0
- {rbms-0.3.2 → rbms-0.4.0}/rbms/potts_bernoulli/classes.py +5 -1
- {rbms-0.3.2 → rbms-0.4.0}/rbms/potts_bernoulli/functional.py +9 -1
- {rbms-0.3.2 → rbms-0.4.0}/rbms/potts_bernoulli/implement.py +12 -0
- {rbms-0.3.2 → rbms-0.4.0}/rbms/scripts/train_rbm.py +9 -7
- {rbms-0.3.2 → rbms-0.4.0}/rbms/training/pcd.py +19 -35
- {rbms-0.3.2 → rbms-0.4.0}/rbms/training/utils.py +61 -9
- {rbms-0.3.2 → rbms-0.4.0}/tests/conftest.py +3 -0
- {rbms-0.3.2 → rbms-0.4.0}/tests/unit_test/bernoulli_bernoulli/test_bernoulli_utils_bbrbm.py +4 -0
- {rbms-0.3.2 → rbms-0.4.0}/tests/unit_test/potts_bernoulli/test_utils_pbrbm.py +4 -0
- {rbms-0.3.2 → rbms-0.4.0}/tests/unit_test/training/test_pcd.py +3 -0
- {rbms-0.3.2 → rbms-0.4.0}/tests/use_cases/test_bbrbm.py +3 -0
- {rbms-0.3.2 → rbms-0.4.0}/tests/use_cases/test_pbrbm.py +6 -0
- rbms-0.3.2/rbms/dataset/__init__.py +0 -100
- {rbms-0.3.2 → rbms-0.4.0}/.github/workflows/codecov.yaml +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/.github/workflows/test.yaml +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/.github/workflows/whats_new.yaml +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/.gitignore +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/LICENSE +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/MANIFEST.in +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/README.md +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/rbms/__init__.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/rbms/bernoulli_bernoulli/__init__.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/rbms/const.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/rbms/custom_fn.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/rbms/dataset/dataset_class.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/rbms/dataset/fasta_utils.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/rbms/dataset/load_fasta.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/rbms/dataset/load_h5.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/rbms/dataset/parser.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/rbms/dataset/utils.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/rbms/map_model.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/rbms/metrics/__init__.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/rbms/metrics/aats.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/rbms/partition_function/ais.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/rbms/partition_function/exact.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/rbms/plot.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/rbms/potts_bernoulli/__init__.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/rbms/potts_bernoulli/tools.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/rbms/potts_bernoulli/utils.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/rbms/sampling/gibbs.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/rbms/sampling/pt.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/rbms/scripts/__init__.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/rbms/scripts/entrypoint.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/rbms/scripts/pt_sampling.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/rbms/utils.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/tests/unit_test/bernoulli_bernoulli/implement/test_compute_energy_bbrbm.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/tests/unit_test/bernoulli_bernoulli/implement/test_compute_energy_hiddens_bbrbm.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/tests/unit_test/bernoulli_bernoulli/implement/test_compute_energy_visibles_bbrbm.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/tests/unit_test/bernoulli_bernoulli/implement/test_compute_gradient_bbrbm.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/tests/unit_test/bernoulli_bernoulli/implement/test_init_chains_bbrbm.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/tests/unit_test/bernoulli_bernoulli/implement/test_init_parameters_bbrbm.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/tests/unit_test/bernoulli_bernoulli/implement/test_sample_hiddens_bbrbm.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/tests/unit_test/bernoulli_bernoulli/implement/test_sample_visibles_bbrbm.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/tests/unit_test/bernoulli_bernoulli/test_functional_bbrbm.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/tests/unit_test/bernoulli_bernoulli/test_methods_bbrbm.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/tests/unit_test/metrics/test_aats.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/tests/unit_test/potts_bernoulli/implement/test_compute_energy_hiddens_pbrbm.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/tests/unit_test/potts_bernoulli/implement/test_compute_energy_pbrbm.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/tests/unit_test/potts_bernoulli/implement/test_compute_energy_visibles_pbrbm.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/tests/unit_test/potts_bernoulli/implement/test_compute_gradient_pbrbm.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/tests/unit_test/potts_bernoulli/implement/test_init_chains_pbrbm.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/tests/unit_test/potts_bernoulli/implement/test_init_parameters_pbrbm.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/tests/unit_test/potts_bernoulli/implement/test_sample_hiddens_pbrbm.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/tests/unit_test/potts_bernoulli/implement/test_sample_visibles_pbrbm.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/tests/unit_test/potts_bernoulli/test_functional_pbrbm.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/tests/unit_test/potts_bernoulli/test_methods_pbrbm.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/tests/unit_test/test_custom_fn.py +0 -0
- {rbms-0.3.2 → rbms-0.4.0}/tests/unit_test/test_utils.py +0 -0
|
@@ -104,7 +104,9 @@ class BBRBM(RBM):
|
|
|
104
104
|
weight_matrix=self.weight_matrix,
|
|
105
105
|
)
|
|
106
106
|
|
|
107
|
-
def compute_gradient(
|
|
107
|
+
def compute_gradient(
|
|
108
|
+
self, data, chains, centered=True, lambda_l1=0.0, lambda_l2=0.0
|
|
109
|
+
):
|
|
108
110
|
_compute_gradient(
|
|
109
111
|
v_data=data["visible"],
|
|
110
112
|
mh_data=data["hidden_mag"],
|
|
@@ -116,6 +118,8 @@ class BBRBM(RBM):
|
|
|
116
118
|
hbias=self.hbias,
|
|
117
119
|
weight_matrix=self.weight_matrix,
|
|
118
120
|
centered=centered,
|
|
121
|
+
lambda_l1=lambda_l1,
|
|
122
|
+
lambda_l2=lambda_l2,
|
|
119
123
|
)
|
|
120
124
|
|
|
121
125
|
def independent_model(self):
|
|
@@ -118,6 +118,8 @@ def compute_gradient(
|
|
|
118
118
|
chains: dict[str, Tensor],
|
|
119
119
|
params: BBRBM,
|
|
120
120
|
centered: bool = True,
|
|
121
|
+
lambda_l1: float = 0.0,
|
|
122
|
+
lambda_l2: float = 0.0,
|
|
121
123
|
) -> None:
|
|
122
124
|
"""Compute the gradient for each of the parameters and attach it.
|
|
123
125
|
|
|
@@ -126,6 +128,8 @@ def compute_gradient(
|
|
|
126
128
|
chains (dict[str, Tensor]): The parallel chains used for gradient computation.
|
|
127
129
|
params (BBRBM): The parameters of the RBM.
|
|
128
130
|
centered (bool, optional): Whether to use centered gradients. Defaults to True.
|
|
131
|
+
lambda_l1 (float, optional): factor for the L1 regularization. Defaults to 0.
|
|
132
|
+
lambda_l2 (float, optional): factor for the L2 regularization. Defaults to 0.
|
|
129
133
|
"""
|
|
130
134
|
_compute_gradient(
|
|
131
135
|
v_data=data["visible"],
|
|
@@ -138,6 +142,8 @@ def compute_gradient(
|
|
|
138
142
|
hbias=params.hbias,
|
|
139
143
|
weight_matrix=params.weight_matrix,
|
|
140
144
|
centered=centered,
|
|
145
|
+
lambda_l1=lambda_l1,
|
|
146
|
+
lambda_l2=lambda_l2,
|
|
141
147
|
)
|
|
142
148
|
|
|
143
149
|
|
|
@@ -169,7 +175,9 @@ def init_chains(
|
|
|
169
175
|
start_v=start_v,
|
|
170
176
|
)
|
|
171
177
|
if weights is None:
|
|
172
|
-
weights = torch.ones(
|
|
178
|
+
weights = torch.ones(
|
|
179
|
+
visible.shape[0], device=visible.device, dtype=visible.dtype
|
|
180
|
+
)
|
|
173
181
|
return dict(
|
|
174
182
|
visible=visible,
|
|
175
183
|
hidden=hidden,
|
|
@@ -77,6 +77,8 @@ def _compute_gradient(
|
|
|
77
77
|
hbias: Tensor,
|
|
78
78
|
weight_matrix: Tensor,
|
|
79
79
|
centered: bool = True,
|
|
80
|
+
lambda_l1: float = 0.0,
|
|
81
|
+
lambda_l2: float = 0.0,
|
|
80
82
|
) -> None:
|
|
81
83
|
w_data = w_data.view(-1, 1)
|
|
82
84
|
w_chain = w_chain.view(-1, 1)
|
|
@@ -118,6 +120,16 @@ def _compute_gradient(
|
|
|
118
120
|
grad_vbias = v_data_mean - v_gen_mean
|
|
119
121
|
grad_hbias = h_data_mean - h_gen_mean
|
|
120
122
|
|
|
123
|
+
if lambda_l1 > 0:
|
|
124
|
+
grad_weight_matrix -= 2 * lambda_l2 * torch.sign(weight_matrix)
|
|
125
|
+
grad_vbias -= 2 * lambda_l2 * torch.sign(vbias)
|
|
126
|
+
grad_hbias -= 2 * lambda_l2 * torch.sign(hbias)
|
|
127
|
+
|
|
128
|
+
if lambda_l2 > 0:
|
|
129
|
+
grad_weight_matrix -= 2 * lambda_l2 * weight_matrix
|
|
130
|
+
grad_vbias -= 2 * lambda_l2 * vbias
|
|
131
|
+
grad_hbias -= 2 * lambda_l2 * hbias
|
|
132
|
+
|
|
121
133
|
# Attach to the parameters
|
|
122
134
|
|
|
123
135
|
weight_matrix.grad.set_(grad_weight_matrix)
|
|
@@ -76,7 +76,12 @@ class EBM(ABC):
|
|
|
76
76
|
|
|
77
77
|
@abstractmethod
|
|
78
78
|
def compute_gradient(
|
|
79
|
-
self,
|
|
79
|
+
self,
|
|
80
|
+
data: dict[str, Tensor],
|
|
81
|
+
chains: dict[str, Tensor],
|
|
82
|
+
centered: bool = True,
|
|
83
|
+
lambda_l1: float = 0.0,
|
|
84
|
+
lambda_l2: float = 0.0,
|
|
80
85
|
) -> None:
|
|
81
86
|
"""Compute the gradient for each of the parameters and attach it.
|
|
82
87
|
|
|
@@ -84,6 +89,8 @@ class EBM(ABC):
|
|
|
84
89
|
data (dict[str, Tensor]): The data state.
|
|
85
90
|
chains (dict[str, Tensor]): The parallel chains used for gradient computation.
|
|
86
91
|
centered (bool, optional): Whether to use centered gradients. Defaults to True.
|
|
92
|
+
lambda_l1 (float, optional): factor for the L1 regularization. Defaults to 0.
|
|
93
|
+
lambda_l2 (float, optional): factor for the L2 regularization. Defaults to 0.
|
|
87
94
|
"""
|
|
88
95
|
...
|
|
89
96
|
|
|
@@ -0,0 +1,83 @@
|
|
|
1
|
+
from pathlib import Path
|
|
2
|
+
from typing import List, Optional, Tuple
|
|
3
|
+
|
|
4
|
+
import numpy as np
|
|
5
|
+
import torch
|
|
6
|
+
|
|
7
|
+
from rbms.dataset.dataset_class import RBMDataset
|
|
8
|
+
from rbms.dataset.load_fasta import load_FASTA
|
|
9
|
+
from rbms.dataset.load_h5 import load_HDF5
|
|
10
|
+
from rbms.dataset.utils import get_subset_labels
|
|
11
|
+
|
|
12
|
+
|
|
13
|
+
def load_dataset(
|
|
14
|
+
dataset_name: str,
|
|
15
|
+
test_dataset_name: Optional[str] = None,
|
|
16
|
+
subset_labels: Optional[List[int]] = None,
|
|
17
|
+
use_weights: bool = False,
|
|
18
|
+
binarize: bool = False,
|
|
19
|
+
alphabet="protein",
|
|
20
|
+
device: str = "cpu",
|
|
21
|
+
dtype: torch.dtype = torch.float32,
|
|
22
|
+
) -> Tuple[RBMDataset, RBMDataset | None]:
|
|
23
|
+
data = None
|
|
24
|
+
weights = None
|
|
25
|
+
names = None
|
|
26
|
+
labels = None
|
|
27
|
+
is_binary = True
|
|
28
|
+
|
|
29
|
+
return_datasets = []
|
|
30
|
+
for dset_name in [dataset_name, test_dataset_name]:
|
|
31
|
+
|
|
32
|
+
if dset_name is not None:
|
|
33
|
+
dataset_name = Path(dataset_name)
|
|
34
|
+
|
|
35
|
+
match dataset_name.suffix:
|
|
36
|
+
case ".h5":
|
|
37
|
+
data, labels = load_HDF5(filename=dataset_name, binarize=binarize)
|
|
38
|
+
case ".fasta":
|
|
39
|
+
data, weights, names = load_FASTA(
|
|
40
|
+
filename=dataset_name,
|
|
41
|
+
binarize=binarize,
|
|
42
|
+
use_weights=use_weights,
|
|
43
|
+
alphabet=alphabet,
|
|
44
|
+
device=device,
|
|
45
|
+
)
|
|
46
|
+
if not binarize:
|
|
47
|
+
is_binary = False
|
|
48
|
+
case _:
|
|
49
|
+
raise ValueError(
|
|
50
|
+
"""
|
|
51
|
+
Dataset could not be loaded as the type is not recognized.
|
|
52
|
+
It should be either:
|
|
53
|
+
- '.h5',
|
|
54
|
+
- '.fasta'
|
|
55
|
+
"""
|
|
56
|
+
)
|
|
57
|
+
|
|
58
|
+
# Select subset of dataset w.r.t. labels
|
|
59
|
+
if subset_labels is not None and labels is not None:
|
|
60
|
+
data, labels = get_subset_labels(data, labels, subset_labels)
|
|
61
|
+
|
|
62
|
+
if weights is None:
|
|
63
|
+
weights = np.ones(data.shape[0])
|
|
64
|
+
if names is None:
|
|
65
|
+
names = np.arange(data.shape[0])
|
|
66
|
+
if labels is None:
|
|
67
|
+
labels = -np.ones(data.shape[0])
|
|
68
|
+
|
|
69
|
+
return_datasets.append(
|
|
70
|
+
RBMDataset(
|
|
71
|
+
data=data,
|
|
72
|
+
labels=labels,
|
|
73
|
+
weights=weights,
|
|
74
|
+
names=names,
|
|
75
|
+
dataset_name=dataset_name,
|
|
76
|
+
is_binary=is_binary,
|
|
77
|
+
device=device,
|
|
78
|
+
dtype=dtype,
|
|
79
|
+
)
|
|
80
|
+
)
|
|
81
|
+
else:
|
|
82
|
+
return_datasets.append(None)
|
|
83
|
+
return tuple(return_datasets)
|
|
@@ -130,6 +130,8 @@ def load_model(
|
|
|
130
130
|
hyperparameters["learning_rate"] = float(
|
|
131
131
|
f["hyperparameters"]["learning_rate"][()]
|
|
132
132
|
)
|
|
133
|
+
hyperparameters["L1"] = float(f["hyperparameters"]["L1"][()])
|
|
134
|
+
hyperparameters["L2"] = float(f["hyperparameters"]["L2"][()])
|
|
133
135
|
if "seed" in f["hyperparameters"].keys():
|
|
134
136
|
hyperparameters["seed"] = int(f["hyperparameters"]["seed"][()])
|
|
135
137
|
if "train_size" in f["hyperparameters"].keys():
|
|
@@ -134,6 +134,29 @@ def add_args_rbm(parser: argparse.ArgumentParser) -> argparse.ArgumentParser:
|
|
|
134
134
|
action="store_true",
|
|
135
135
|
help="(Defaults to False). Restore the training",
|
|
136
136
|
)
|
|
137
|
+
rbm_args.add_argument(
|
|
138
|
+
"--no_center",
|
|
139
|
+
default=False,
|
|
140
|
+
action="store_true",
|
|
141
|
+
help="(Defaults to False). Use the non-centered gradient.",
|
|
142
|
+
)
|
|
143
|
+
return parser
|
|
144
|
+
|
|
145
|
+
|
|
146
|
+
def add_args_regularization(parser: argparse.ArgumentParser) -> argparse.ArgumentParser:
|
|
147
|
+
reg_args = parser.add_argument_group("Regularization")
|
|
148
|
+
reg_args.add_argument(
|
|
149
|
+
"--L1",
|
|
150
|
+
default=None,
|
|
151
|
+
type=float,
|
|
152
|
+
help="(Defaults to 0.0). Lambda parameter for the L1 regularization.",
|
|
153
|
+
)
|
|
154
|
+
reg_args.add_argument(
|
|
155
|
+
"--L2",
|
|
156
|
+
default=None,
|
|
157
|
+
type=float,
|
|
158
|
+
help="(Defaults to 0.0). Lambda parameter for the L2 regularization.",
|
|
159
|
+
)
|
|
137
160
|
return parser
|
|
138
161
|
|
|
139
162
|
|
|
@@ -186,6 +209,9 @@ default_args: dict[str, Any] = {
|
|
|
186
209
|
"beta": 1.0,
|
|
187
210
|
"restore": False,
|
|
188
211
|
"seed": np.random.randint(0, 1000000000000),
|
|
212
|
+
"no_center": False,
|
|
213
|
+
"L1": 0.0,
|
|
214
|
+
"L2": 0.0,
|
|
189
215
|
}
|
|
190
216
|
|
|
191
217
|
|
|
@@ -105,7 +105,9 @@ class PBRBM(RBM):
|
|
|
105
105
|
weight_matrix=self.weight_matrix,
|
|
106
106
|
)
|
|
107
107
|
|
|
108
|
-
def compute_gradient(
|
|
108
|
+
def compute_gradient(
|
|
109
|
+
self, data, chains, centered=True, lambda_l1=0.0, lambda_l2=0.0
|
|
110
|
+
):
|
|
109
111
|
_compute_gradient(
|
|
110
112
|
v_data=data["visible"],
|
|
111
113
|
mh_data=data["hidden_mag"],
|
|
@@ -117,6 +119,8 @@ class PBRBM(RBM):
|
|
|
117
119
|
hbias=self.hbias,
|
|
118
120
|
weight_matrix=self.weight_matrix,
|
|
119
121
|
centered=centered,
|
|
122
|
+
lambda_l1=0.0,
|
|
123
|
+
lambda_l2=lambda_l2,
|
|
120
124
|
)
|
|
121
125
|
|
|
122
126
|
def independent_model(self):
|
|
@@ -118,6 +118,8 @@ def compute_gradient(
|
|
|
118
118
|
chains: dict[str, Tensor],
|
|
119
119
|
params: PBRBM,
|
|
120
120
|
centered: bool = True,
|
|
121
|
+
lambda_l1: float = 0.0,
|
|
122
|
+
lambda_l2: float = 0.0,
|
|
121
123
|
) -> None:
|
|
122
124
|
"""Compute the gradient for each of the parameters and attach it.
|
|
123
125
|
|
|
@@ -126,6 +128,8 @@ def compute_gradient(
|
|
|
126
128
|
chains (dict[str, Tensor]): The parallel chains used for gradient computation.
|
|
127
129
|
params (PBRBM): The parameters of the RBM.
|
|
128
130
|
centered (bool, optional): Whether to use centered gradients. Defaults to True.
|
|
131
|
+
lambda_l1 (float, optional): factor for the L1 regularization. Defaults to 0.
|
|
132
|
+
lambda_l2 (float, optional): factor for the L2 regularization. Defaults to 0.
|
|
129
133
|
"""
|
|
130
134
|
_compute_gradient(
|
|
131
135
|
v_data=data["visible"],
|
|
@@ -138,6 +142,8 @@ def compute_gradient(
|
|
|
138
142
|
hbias=params.hbias,
|
|
139
143
|
weight_matrix=params.weight_matrix,
|
|
140
144
|
centered=centered,
|
|
145
|
+
lambda_l1=lambda_l1,
|
|
146
|
+
lambda_l2=lambda_l2,
|
|
141
147
|
)
|
|
142
148
|
|
|
143
149
|
|
|
@@ -169,7 +175,9 @@ def init_chains(
|
|
|
169
175
|
start_v=start_v,
|
|
170
176
|
)
|
|
171
177
|
if weights is None:
|
|
172
|
-
weights = torch.ones(
|
|
178
|
+
weights = torch.ones(
|
|
179
|
+
visible.shape[0], device=visible.device, dtype=visible.dtype
|
|
180
|
+
)
|
|
173
181
|
return dict(
|
|
174
182
|
visible=visible,
|
|
175
183
|
hidden=hidden,
|
|
@@ -97,6 +97,8 @@ def _compute_gradient(
|
|
|
97
97
|
hbias: Tensor,
|
|
98
98
|
weight_matrix: Tensor,
|
|
99
99
|
centered: bool = True,
|
|
100
|
+
lambda_l1: float = 0.0,
|
|
101
|
+
lambda_l2: float = 0.0,
|
|
100
102
|
):
|
|
101
103
|
w_data = w_data.view(-1, 1, 1)
|
|
102
104
|
w_chain = w_chain.view(-1, 1, 1)
|
|
@@ -175,6 +177,16 @@ def _compute_gradient(
|
|
|
175
177
|
|
|
176
178
|
grad_vbias = v_data_mean - v_gen_mean
|
|
177
179
|
grad_hbias = h_data_mean - h_gen_mean
|
|
180
|
+
|
|
181
|
+
if lambda_l1 > 0:
|
|
182
|
+
grad_weight_matrix -= 2 * lambda_l2 * torch.sign(weight_matrix)
|
|
183
|
+
grad_vbias -= 2 * lambda_l2 * torch.sign(vbias)
|
|
184
|
+
grad_hbias -= 2 * lambda_l2 * torch.sign(hbias)
|
|
185
|
+
|
|
186
|
+
if lambda_l2 != 0:
|
|
187
|
+
grad_weight_matrix -= 2 * lambda_l2 * weight_matrix
|
|
188
|
+
grad_vbias -= 2 * lambda_l2 * vbias
|
|
189
|
+
grad_hbias -= 2 * lambda_l2 * hbias
|
|
178
190
|
weight_matrix.grad.set_(grad_weight_matrix)
|
|
179
191
|
vbias.grad.set_(grad_vbias)
|
|
180
192
|
hbias.grad.set_(grad_hbias)
|
|
@@ -9,6 +9,7 @@ from rbms.parser import (
|
|
|
9
9
|
add_args_pytorch,
|
|
10
10
|
add_args_rbm,
|
|
11
11
|
add_args_saves,
|
|
12
|
+
add_args_regularization,
|
|
12
13
|
match_args_dtype,
|
|
13
14
|
remove_argument,
|
|
14
15
|
default_args,
|
|
@@ -21,6 +22,7 @@ def create_parser():
|
|
|
21
22
|
parser = argparse.ArgumentParser(description="Train a Restricted Boltzmann Machine")
|
|
22
23
|
parser = add_args_dataset(parser)
|
|
23
24
|
parser = add_args_rbm(parser)
|
|
25
|
+
parser = add_args_regularization(parser)
|
|
24
26
|
parser = add_args_saves(parser)
|
|
25
27
|
parser = add_args_pytorch(parser)
|
|
26
28
|
remove_argument(parser, "use_torch")
|
|
@@ -28,28 +30,28 @@ def create_parser():
|
|
|
28
30
|
|
|
29
31
|
|
|
30
32
|
def train_rbm(args: dict):
|
|
33
|
+
if args["num_updates"] is None:
|
|
34
|
+
args["num_updates"] = default_args["num_updates"]
|
|
31
35
|
checkpoints = get_checkpoints(
|
|
32
36
|
num_updates=args["num_updates"], n_save=args["n_save"], spacing=args["spacing"]
|
|
33
37
|
)
|
|
34
|
-
|
|
38
|
+
train_dataset, test_dataset = load_dataset(
|
|
35
39
|
dataset_name=args["data"],
|
|
36
40
|
subset_labels=args["subset_labels"],
|
|
37
41
|
use_weights=args["use_weights"],
|
|
38
42
|
alphabet=args["alphabet"],
|
|
39
43
|
binarize=args["binarize"],
|
|
40
|
-
train_size=1.0,
|
|
41
|
-
test_size=None,
|
|
42
44
|
device=args["device"],
|
|
43
45
|
dtype=args["dtype"],
|
|
44
|
-
seed=args["seed"],
|
|
45
46
|
)
|
|
46
|
-
print(
|
|
47
|
-
if
|
|
47
|
+
print(train_dataset)
|
|
48
|
+
if train_dataset.is_binary:
|
|
48
49
|
model_type = "BBRBM"
|
|
49
50
|
else:
|
|
50
51
|
model_type = "PBRBM"
|
|
51
52
|
train(
|
|
52
|
-
|
|
53
|
+
train_dataset=train_dataset,
|
|
54
|
+
test_dataset=test_dataset,
|
|
53
55
|
model_type=model_type,
|
|
54
56
|
args=args,
|
|
55
57
|
dtype=args["dtype"],
|
|
@@ -1,5 +1,5 @@
|
|
|
1
1
|
import time
|
|
2
|
-
from typing import Tuple
|
|
2
|
+
from typing import Tuple, Optional
|
|
3
3
|
|
|
4
4
|
import numpy as np
|
|
5
5
|
import torch
|
|
@@ -12,7 +12,7 @@ from rbms.io import save_model
|
|
|
12
12
|
from rbms.map_model import map_model
|
|
13
13
|
from rbms.potts_bernoulli.classes import PBRBM
|
|
14
14
|
from rbms.potts_bernoulli.utils import ensure_zero_sum_gauge
|
|
15
|
-
from rbms.training.utils import
|
|
15
|
+
from rbms.training.utils import setup_training, initialize_model_archive
|
|
16
16
|
from rbms.utils import check_file_existence, log_to_csv
|
|
17
17
|
from rbms.parser import default_args, set_args_default
|
|
18
18
|
|
|
@@ -54,7 +54,8 @@ def fit_batch_pcd(
|
|
|
54
54
|
|
|
55
55
|
|
|
56
56
|
def train(
|
|
57
|
-
|
|
57
|
+
train_dataset: RBMDataset,
|
|
58
|
+
test_dataset: Optional[RBMDataset],
|
|
58
59
|
model_type: str,
|
|
59
60
|
args: dict,
|
|
60
61
|
dtype: torch.dtype,
|
|
@@ -73,41 +74,18 @@ def train(
|
|
|
73
74
|
checkpoints (np.ndarray): An array of checkpoints for saving model states.
|
|
74
75
|
"""
|
|
75
76
|
|
|
76
|
-
filename = args["filename"]
|
|
77
77
|
if not (args["overwrite"]):
|
|
78
|
-
check_file_existence(filename)
|
|
78
|
+
check_file_existence(args["filename"])
|
|
79
79
|
|
|
80
|
-
num_visibles = dataset.get_num_visibles()
|
|
81
80
|
# Create a first archive with the initialized model
|
|
82
81
|
if not (args["restore"]):
|
|
83
|
-
|
|
84
|
-
|
|
85
|
-
|
|
86
|
-
|
|
87
|
-
|
|
88
|
-
params = map_model[model_type].init_parameters(
|
|
89
|
-
num_hiddens=args["num_hiddens"],
|
|
90
|
-
dataset=train_dataset,
|
|
91
|
-
device=args["device"],
|
|
82
|
+
initialize_model_archive(
|
|
83
|
+
args=args,
|
|
84
|
+
model_type=model_type,
|
|
85
|
+
train_dataset=train_dataset,
|
|
86
|
+
test_dataset=test_dataset,
|
|
92
87
|
dtype=dtype,
|
|
93
88
|
)
|
|
94
|
-
|
|
95
|
-
if isinstance(params, PBRBM):
|
|
96
|
-
ensure_zero_sum_gauge(params)
|
|
97
|
-
create_machine(
|
|
98
|
-
filename=filename,
|
|
99
|
-
params=params,
|
|
100
|
-
num_visibles=num_visibles,
|
|
101
|
-
num_hiddens=args["num_hiddens"],
|
|
102
|
-
num_chains=args["num_chains"],
|
|
103
|
-
batch_size=args["batch_size"],
|
|
104
|
-
gibbs_steps=args["gibbs_steps"],
|
|
105
|
-
learning_rate=args["learning_rate"],
|
|
106
|
-
train_size=args["train_size"],
|
|
107
|
-
log=args["log"],
|
|
108
|
-
flags=["checkpoint"],
|
|
109
|
-
seed=args["seed"],
|
|
110
|
-
)
|
|
111
89
|
(
|
|
112
90
|
params,
|
|
113
91
|
parallel_chains,
|
|
@@ -119,7 +97,12 @@ def train(
|
|
|
119
97
|
pbar,
|
|
120
98
|
train_dataset,
|
|
121
99
|
test_dataset,
|
|
122
|
-
) = setup_training(
|
|
100
|
+
) = setup_training(
|
|
101
|
+
args,
|
|
102
|
+
map_model=map_model,
|
|
103
|
+
train_dataset=train_dataset,
|
|
104
|
+
test_dataset=test_dataset,
|
|
105
|
+
)
|
|
123
106
|
args = set_args_default(args=args, default_args=default_args)
|
|
124
107
|
optimizer = SGD(params.parameters(), lr=args["learning_rate"], maximize=True)
|
|
125
108
|
|
|
@@ -129,8 +112,8 @@ def train(
|
|
|
129
112
|
# Continue the training
|
|
130
113
|
with torch.no_grad():
|
|
131
114
|
for idx in range(num_updates + 1, args["num_updates"] + 1):
|
|
132
|
-
rand_idx = torch.randperm(len(
|
|
133
|
-
batch = (
|
|
115
|
+
rand_idx = torch.randperm(len(train_dataset))[: args["batch_size"]]
|
|
116
|
+
batch = (train_dataset.data[rand_idx], train_dataset.weights[rand_idx])
|
|
134
117
|
|
|
135
118
|
optimizer.zero_grad(set_to_none=False)
|
|
136
119
|
parallel_chains, logs = fit_batch_pcd(
|
|
@@ -139,6 +122,7 @@ def train(
|
|
|
139
122
|
params=params,
|
|
140
123
|
gibbs_steps=args["gibbs_steps"],
|
|
141
124
|
beta=args["beta"],
|
|
125
|
+
centered=not (args["no_center"]),
|
|
142
126
|
)
|
|
143
127
|
optimizer.step()
|
|
144
128
|
if isinstance(params, PBRBM):
|
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
import pathlib
|
|
2
2
|
import time
|
|
3
|
-
from typing import Any, List, Tuple
|
|
3
|
+
from typing import Any, List, Optional, Tuple
|
|
4
4
|
|
|
5
5
|
import h5py
|
|
6
6
|
import numpy as np
|
|
@@ -10,15 +10,19 @@ from tqdm import tqdm
|
|
|
10
10
|
|
|
11
11
|
from rbms.classes import EBM
|
|
12
12
|
from rbms.const import LOG_FILE_HEADER
|
|
13
|
+
from rbms.dataset.dataset_class import RBMDataset
|
|
13
14
|
from rbms.io import load_model, save_model
|
|
14
15
|
from rbms.map_model import map_model
|
|
16
|
+
from rbms.parser import default_args, set_args_default
|
|
17
|
+
from rbms.potts_bernoulli.classes import PBRBM
|
|
18
|
+
from rbms.potts_bernoulli.utils import ensure_zero_sum_gauge
|
|
15
19
|
from rbms.utils import get_saved_updates
|
|
16
|
-
from rbms.dataset.dataset_class import RBMDataset
|
|
17
20
|
|
|
18
21
|
|
|
19
22
|
def setup_training(
|
|
20
23
|
args: dict,
|
|
21
|
-
|
|
24
|
+
train_dataset: RBMDataset,
|
|
25
|
+
test_dataset: Optional[RBMDataset] = None,
|
|
22
26
|
map_model: dict[str, EBM] = map_model,
|
|
23
27
|
) -> Tuple[
|
|
24
28
|
EBM,
|
|
@@ -52,14 +56,15 @@ def setup_training(
|
|
|
52
56
|
|
|
53
57
|
# Hyperparameters
|
|
54
58
|
for k, v in hyperparameters.items():
|
|
55
|
-
if
|
|
59
|
+
if args[k] is None:
|
|
56
60
|
args[k] = v
|
|
57
61
|
|
|
58
|
-
|
|
59
|
-
|
|
60
|
-
|
|
61
|
-
|
|
62
|
-
|
|
62
|
+
if test_dataset is None:
|
|
63
|
+
train_dataset, test_dataset = train_dataset.split_train_test(
|
|
64
|
+
rng=np.random.default_rng(args["seed"]),
|
|
65
|
+
train_size=args["train_size"],
|
|
66
|
+
test_size=args["test_size"],
|
|
67
|
+
)
|
|
63
68
|
|
|
64
69
|
# Open the log file if it exists
|
|
65
70
|
log_filename = pathlib.Path(args["filename"]).parent / pathlib.Path(
|
|
@@ -111,6 +116,8 @@ def create_machine(
|
|
|
111
116
|
log: bool,
|
|
112
117
|
flags: List[str],
|
|
113
118
|
seed: int,
|
|
119
|
+
L1: float,
|
|
120
|
+
L2: float,
|
|
114
121
|
) -> None:
|
|
115
122
|
"""Create a RBM and save it to a new file.
|
|
116
123
|
|
|
@@ -124,6 +131,8 @@ def create_machine(
|
|
|
124
131
|
gibbs_steps (int): Number of Gibbs steps to perform.
|
|
125
132
|
learning_rate (float): Learning rate for training.
|
|
126
133
|
log (bool): Whether to enable logging.
|
|
134
|
+
L1 (float): Lambda parameter for L1 regularization.
|
|
135
|
+
L2 (float): Lambda parameter for L2 regularization.
|
|
127
136
|
"""
|
|
128
137
|
# Permanent chains
|
|
129
138
|
parallel_chains = params.init_chains(num_samples=num_chains)
|
|
@@ -139,6 +148,8 @@ def create_machine(
|
|
|
139
148
|
hyperparameters["learning_rate"] = learning_rate
|
|
140
149
|
hyperparameters["train_size"] = train_size
|
|
141
150
|
hyperparameters["seed"] = seed
|
|
151
|
+
hyperparameters["L1"] = L1
|
|
152
|
+
hyperparameters["L2"] = L2
|
|
142
153
|
|
|
143
154
|
save_model(
|
|
144
155
|
filename=filename,
|
|
@@ -182,3 +193,44 @@ def get_checkpoints(num_updates: int, n_save: int, spacing: str = "exp") -> np.n
|
|
|
182
193
|
)
|
|
183
194
|
checkpoints = np.unique(np.append(checkpoints, num_updates))
|
|
184
195
|
return checkpoints
|
|
196
|
+
|
|
197
|
+
|
|
198
|
+
def initialize_model_archive(
|
|
199
|
+
args: dict,
|
|
200
|
+
model_type: str,
|
|
201
|
+
train_dataset: RBMDataset,
|
|
202
|
+
test_dataset: Optional[RBMDataset],
|
|
203
|
+
dtype: torch.dtype,
|
|
204
|
+
):
|
|
205
|
+
num_visibles = train_dataset.get_num_visibles()
|
|
206
|
+
args = set_args_default(args=args, default_args=default_args)
|
|
207
|
+
rng = np.random.default_rng(args["seed"])
|
|
208
|
+
if test_dataset is None:
|
|
209
|
+
train_dataset, _ = train_dataset.split_train_test(
|
|
210
|
+
rng, args["train_size"], args["test_size"]
|
|
211
|
+
)
|
|
212
|
+
params = map_model[model_type].init_parameters(
|
|
213
|
+
num_hiddens=args["num_hiddens"],
|
|
214
|
+
dataset=train_dataset,
|
|
215
|
+
device=args["device"],
|
|
216
|
+
dtype=dtype,
|
|
217
|
+
)
|
|
218
|
+
|
|
219
|
+
if isinstance(params, PBRBM):
|
|
220
|
+
ensure_zero_sum_gauge(params)
|
|
221
|
+
create_machine(
|
|
222
|
+
filename=args["filename"],
|
|
223
|
+
params=params,
|
|
224
|
+
num_visibles=num_visibles,
|
|
225
|
+
num_hiddens=args["num_hiddens"],
|
|
226
|
+
num_chains=args["num_chains"],
|
|
227
|
+
batch_size=args["batch_size"],
|
|
228
|
+
gibbs_steps=args["gibbs_steps"],
|
|
229
|
+
learning_rate=args["learning_rate"],
|
|
230
|
+
train_size=args["train_size"],
|
|
231
|
+
log=args["log"],
|
|
232
|
+
flags=["checkpoint"],
|
|
233
|
+
seed=args["seed"],
|
|
234
|
+
L1=args["L1"],
|
|
235
|
+
L2=args["L2"],
|
|
236
|
+
)
|
|
@@ -39,6 +39,8 @@ def test_create_load_machine(tmp_path, sample_params_class_bbrbm):
|
|
|
39
39
|
log=True,
|
|
40
40
|
flags=["test"],
|
|
41
41
|
seed=pytest.SEED,
|
|
42
|
+
L1=1.0,
|
|
43
|
+
L2=0.0,
|
|
42
44
|
)
|
|
43
45
|
|
|
44
46
|
# Check if the file was created
|
|
@@ -87,6 +89,8 @@ def test_create_load_machine_dtype(tmp_path, sample_params_class_bbrbm):
|
|
|
87
89
|
log=True,
|
|
88
90
|
flags=["test"],
|
|
89
91
|
seed=pytest.SEED,
|
|
92
|
+
L1=0.0,
|
|
93
|
+
L2=1.0,
|
|
90
94
|
)
|
|
91
95
|
|
|
92
96
|
# Check if the file was created
|
|
@@ -35,6 +35,8 @@ def test_create_load_machine(tmp_path, sample_params_class_pbrbm):
|
|
|
35
35
|
log=True,
|
|
36
36
|
flags=["test"],
|
|
37
37
|
seed=pytest.SEED,
|
|
38
|
+
L1=1.0,
|
|
39
|
+
L2=0.0,
|
|
38
40
|
)
|
|
39
41
|
|
|
40
42
|
# Check if the file was created
|
|
@@ -83,6 +85,8 @@ def test_create_load_machine_dtype(tmp_path, sample_params_class_pbrbm):
|
|
|
83
85
|
log=True,
|
|
84
86
|
flags=["test"],
|
|
85
87
|
seed=pytest.SEED,
|
|
88
|
+
L1=0.0,
|
|
89
|
+
L2=1.0,
|
|
86
90
|
)
|
|
87
91
|
|
|
88
92
|
# Check if the file was created
|
|
@@ -18,6 +18,7 @@ def test_train_bbrbm(sample_dataset_bbrbm, sample_args):
|
|
|
18
18
|
model_type = "BBRBM"
|
|
19
19
|
train(
|
|
20
20
|
dataset,
|
|
21
|
+
test_dataset,
|
|
21
22
|
model_type,
|
|
22
23
|
sample_args,
|
|
23
24
|
torch.float32,
|
|
@@ -54,12 +55,14 @@ def test_train_bbrbm(sample_dataset_bbrbm, sample_args):
|
|
|
54
55
|
|
|
55
56
|
def test_train_pbrbm(sample_dataset_pbrbm, sample_args):
|
|
56
57
|
dataset = sample_dataset_pbrbm
|
|
58
|
+
test_dataset = sample_dataset_pbrbm
|
|
57
59
|
checkpoints = np.arange(1, sample_args["num_updates"] + 1)
|
|
58
60
|
sample_args["restore"] = False
|
|
59
61
|
sample_args["batch_size"] = pytest.NUM_SAMPLES
|
|
60
62
|
model_type = "PBRBM"
|
|
61
63
|
train(
|
|
62
64
|
dataset,
|
|
65
|
+
test_dataset,
|
|
63
66
|
model_type,
|
|
64
67
|
sample_args,
|
|
65
68
|
torch.float32,
|
|
@@ -57,6 +57,9 @@ def test_use_case_train_pbrbm_no_weights():
|
|
|
57
57
|
"binarize": False,
|
|
58
58
|
"overwrite": True,
|
|
59
59
|
"seed": 42,
|
|
60
|
+
"no_center": False,
|
|
61
|
+
"L1": 0.0,
|
|
62
|
+
"L2": 1.0,
|
|
60
63
|
}
|
|
61
64
|
train_rbm(args)
|
|
62
65
|
|
|
@@ -130,6 +133,9 @@ def test_use_case_train_pbrbm_weights():
|
|
|
130
133
|
"binarize": False,
|
|
131
134
|
"overwrite": True,
|
|
132
135
|
"seed": 42,
|
|
136
|
+
"no_center": False,
|
|
137
|
+
"L1": 1.0,
|
|
138
|
+
"L2": 0.0,
|
|
133
139
|
}
|
|
134
140
|
train_rbm(args)
|
|
135
141
|
|
|
@@ -1,100 +0,0 @@
|
|
|
1
|
-
from pathlib import Path
|
|
2
|
-
from typing import List, Optional
|
|
3
|
-
|
|
4
|
-
import numpy as np
|
|
5
|
-
import torch
|
|
6
|
-
|
|
7
|
-
from rbms.dataset.dataset_class import RBMDataset
|
|
8
|
-
from rbms.dataset.load_fasta import load_FASTA
|
|
9
|
-
from rbms.dataset.load_h5 import load_HDF5
|
|
10
|
-
from rbms.dataset.utils import get_subset_labels
|
|
11
|
-
|
|
12
|
-
|
|
13
|
-
def load_dataset(
|
|
14
|
-
dataset_name: str,
|
|
15
|
-
subset_labels: Optional[List[int]] = None,
|
|
16
|
-
use_weights: bool = False,
|
|
17
|
-
train_size: float = 0.6,
|
|
18
|
-
test_size: Optional[float] = None,
|
|
19
|
-
binarize: bool = False,
|
|
20
|
-
alphabet="protein",
|
|
21
|
-
seed: int = 19023741073419046239412739401234901,
|
|
22
|
-
device: str = "cpu",
|
|
23
|
-
dtype: torch.dtype = torch.float32,
|
|
24
|
-
):
|
|
25
|
-
rng = np.random.default_rng(seed)
|
|
26
|
-
data = None
|
|
27
|
-
weights = None
|
|
28
|
-
names = None
|
|
29
|
-
labels = None
|
|
30
|
-
is_binary = True
|
|
31
|
-
|
|
32
|
-
dataset_name = Path(dataset_name)
|
|
33
|
-
|
|
34
|
-
match dataset_name.suffix:
|
|
35
|
-
case ".h5":
|
|
36
|
-
data, labels = load_HDF5(filename=dataset_name, binarize=binarize)
|
|
37
|
-
case ".fasta":
|
|
38
|
-
data, weights, names = load_FASTA(
|
|
39
|
-
filename=dataset_name,
|
|
40
|
-
binarize=binarize,
|
|
41
|
-
use_weights=use_weights,
|
|
42
|
-
alphabet=alphabet,
|
|
43
|
-
device=device,
|
|
44
|
-
)
|
|
45
|
-
if not binarize:
|
|
46
|
-
is_binary = False
|
|
47
|
-
case _:
|
|
48
|
-
raise ValueError(
|
|
49
|
-
"""
|
|
50
|
-
Dataset could not be loaded as the type is not recognized.
|
|
51
|
-
It should be either:
|
|
52
|
-
- '.h5',
|
|
53
|
-
- '.fasta'
|
|
54
|
-
"""
|
|
55
|
-
)
|
|
56
|
-
|
|
57
|
-
# Select subset of dataset w.r.t. labels
|
|
58
|
-
if subset_labels is not None and labels is not None:
|
|
59
|
-
data, labels = get_subset_labels(data, labels, subset_labels)
|
|
60
|
-
|
|
61
|
-
if weights is None:
|
|
62
|
-
weights = np.ones(data.shape[0])
|
|
63
|
-
if names is None:
|
|
64
|
-
names = np.arange(data.shape[0])
|
|
65
|
-
if labels is None:
|
|
66
|
-
labels = -np.ones(data.shape[0])
|
|
67
|
-
|
|
68
|
-
# Shuffle dataset
|
|
69
|
-
permutation_index = rng.permutation(data.shape[0])
|
|
70
|
-
|
|
71
|
-
# Split train/test
|
|
72
|
-
train_size = int(train_size * data.shape[0])
|
|
73
|
-
if test_size is not None:
|
|
74
|
-
test_size = int(test_size * data.shape[0])
|
|
75
|
-
else:
|
|
76
|
-
test_size = data.shape[0] - train_size
|
|
77
|
-
|
|
78
|
-
train_dataset = RBMDataset(
|
|
79
|
-
data=data[permutation_index[:train_size]],
|
|
80
|
-
labels=labels[permutation_index[:train_size]],
|
|
81
|
-
weights=weights[permutation_index[:train_size]],
|
|
82
|
-
names=names[permutation_index[:train_size]],
|
|
83
|
-
dataset_name=dataset_name,
|
|
84
|
-
is_binary=is_binary,
|
|
85
|
-
device=device,
|
|
86
|
-
dtype=dtype,
|
|
87
|
-
)
|
|
88
|
-
test_dataset = None
|
|
89
|
-
if test_size > 0:
|
|
90
|
-
test_dataset = RBMDataset(
|
|
91
|
-
data=data[permutation_index[train_size : train_size + test_size]],
|
|
92
|
-
labels=labels[permutation_index[train_size : train_size + test_size]],
|
|
93
|
-
weights=weights[permutation_index[train_size : train_size + test_size]],
|
|
94
|
-
names=names[permutation_index[train_size : train_size + test_size]],
|
|
95
|
-
dataset_name=dataset_name,
|
|
96
|
-
is_binary=is_binary,
|
|
97
|
-
device=device,
|
|
98
|
-
dtype=dtype,
|
|
99
|
-
)
|
|
100
|
-
return train_dataset, test_dataset
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
{rbms-0.3.2 → rbms-0.4.0}/tests/unit_test/bernoulli_bernoulli/implement/test_compute_energy_bbrbm.py
RENAMED
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
{rbms-0.3.2 → rbms-0.4.0}/tests/unit_test/bernoulli_bernoulli/implement/test_init_chains_bbrbm.py
RENAMED
|
File without changes
|
|
File without changes
|
{rbms-0.3.2 → rbms-0.4.0}/tests/unit_test/bernoulli_bernoulli/implement/test_sample_hiddens_bbrbm.py
RENAMED
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
{rbms-0.3.2 → rbms-0.4.0}/tests/unit_test/potts_bernoulli/implement/test_compute_energy_pbrbm.py
RENAMED
|
File without changes
|
|
File without changes
|
{rbms-0.3.2 → rbms-0.4.0}/tests/unit_test/potts_bernoulli/implement/test_compute_gradient_pbrbm.py
RENAMED
|
File without changes
|
{rbms-0.3.2 → rbms-0.4.0}/tests/unit_test/potts_bernoulli/implement/test_init_chains_pbrbm.py
RENAMED
|
File without changes
|
{rbms-0.3.2 → rbms-0.4.0}/tests/unit_test/potts_bernoulli/implement/test_init_parameters_pbrbm.py
RENAMED
|
File without changes
|
{rbms-0.3.2 → rbms-0.4.0}/tests/unit_test/potts_bernoulli/implement/test_sample_hiddens_pbrbm.py
RENAMED
|
File without changes
|
{rbms-0.3.2 → rbms-0.4.0}/tests/unit_test/potts_bernoulli/implement/test_sample_visibles_pbrbm.py
RENAMED
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|