iapytoo 0.0.4.dev3__py3-none-any.whl → 0.0.6.dev2__py3-none-any.whl
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.
Potentially problematic release.
This version of iapytoo might be problematic. Click here for more details.
- dockers/docker-compose.yml +15 -0
- dockers/requirements.txt +6 -0
- dockers/run-python.sh +1 -2
- examples/zipped_runs/selected_runs.zip +0 -0
- iapytoo/_version.py +9 -4
- iapytoo/dataset/__init__.py +4 -8
- iapytoo/metrics/__init__.py +0 -82
- iapytoo/metrics/collection.py +39 -0
- iapytoo/metrics/metric.py +16 -9
- iapytoo/metrics/predefined.py +13 -12
- iapytoo/predictions/__init__.py +33 -42
- iapytoo/predictions/feature.py +1 -2
- iapytoo/predictions/plotters.py +39 -11
- iapytoo/predictions/predictors.py +0 -2
- iapytoo/predictions/types.py +0 -2
- iapytoo/train/checkpoint.py +18 -9
- iapytoo/train/factories.py +188 -112
- iapytoo/train/inference.py +175 -0
- iapytoo/train/logger.py +52 -54
- iapytoo/train/loss.py +4 -5
- iapytoo/train/mlflow_model.py +88 -0
- iapytoo/train/training.py +61 -101
- iapytoo/train/valuator.py +61 -0
- iapytoo/train/wgan.py +104 -46
- iapytoo/utils/arguments.py +19 -34
- iapytoo/utils/cf/config.yml +21 -0
- iapytoo/utils/config.py +300 -79
- iapytoo/utils/model_config.py +74 -0
- iapytoo/utils/singleton.py +15 -0
- {iapytoo-0.0.4.dev3.dist-info → iapytoo-0.0.6.dev2.dist-info}/METADATA +3 -2
- iapytoo-0.0.6.dev2.dist-info/RECORD +50 -0
- {iapytoo-0.0.4.dev3.dist-info → iapytoo-0.0.6.dev2.dist-info}/WHEEL +1 -1
- {iapytoo-0.0.4.dev3.dist-info → iapytoo-0.0.6.dev2.dist-info}/top_level.txt +2 -0
- mlflow/relocate_mlruns.py +67 -0
- mlflow/zip_selected_runs.py +80 -0
- iapytoo/utils/meta_singleton.py +0 -17
- iapytoo-0.0.4.dev3.dist-info/RECORD +0 -39
- {iapytoo-0.0.4.dev3.dist-info → iapytoo-0.0.6.dev2.dist-info/licenses}/LICENSE +0 -0
dockers/requirements.txt
ADDED
dockers/run-python.sh
CHANGED
|
Binary file
|
iapytoo/_version.py
CHANGED
|
@@ -1,8 +1,13 @@
|
|
|
1
|
-
# file generated by
|
|
1
|
+
# file generated by setuptools-scm
|
|
2
2
|
# don't change, don't track in version control
|
|
3
|
+
|
|
4
|
+
__all__ = ["__version__", "__version_tuple__", "version", "version_tuple"]
|
|
5
|
+
|
|
3
6
|
TYPE_CHECKING = False
|
|
4
7
|
if TYPE_CHECKING:
|
|
5
|
-
from typing import Tuple
|
|
8
|
+
from typing import Tuple
|
|
9
|
+
from typing import Union
|
|
10
|
+
|
|
6
11
|
VERSION_TUPLE = Tuple[Union[int, str], ...]
|
|
7
12
|
else:
|
|
8
13
|
VERSION_TUPLE = object
|
|
@@ -12,5 +17,5 @@ __version__: str
|
|
|
12
17
|
__version_tuple__: VERSION_TUPLE
|
|
13
18
|
version_tuple: VERSION_TUPLE
|
|
14
19
|
|
|
15
|
-
__version__ = version = '0.0.
|
|
16
|
-
__version_tuple__ = version_tuple = (0, 0,
|
|
20
|
+
__version__ = version = '0.0.6.dev2'
|
|
21
|
+
__version_tuple__ = version_tuple = (0, 0, 6, 'dev2')
|
iapytoo/dataset/__init__.py
CHANGED
|
@@ -13,8 +13,7 @@ class DummyVisionDataset(VisionDataset):
|
|
|
13
13
|
np.random.seed(12)
|
|
14
14
|
A = np.array(range(self.dim * self.dim)).reshape((self.dim, self.dim))
|
|
15
15
|
self.images = np.zeros(
|
|
16
|
-
shape=(self.length, self.n_channels,
|
|
17
|
-
self.dim, self.dim), dtype=np.float32
|
|
16
|
+
shape=(self.length, self.n_channels, self.dim, self.dim), dtype=np.float32
|
|
18
17
|
)
|
|
19
18
|
for idx in range(self.length):
|
|
20
19
|
for c in range(self.n_channels):
|
|
@@ -24,10 +23,8 @@ class DummyVisionDataset(VisionDataset):
|
|
|
24
23
|
|
|
25
24
|
self.targets = np.vstack(
|
|
26
25
|
[
|
|
27
|
-
np.array(range(self.length)) +
|
|
28
|
-
np.random.normal(size=(self.length)),
|
|
29
|
-
np.array(range(self.length)) +
|
|
30
|
-
np.random.normal(size=(self.length)),
|
|
26
|
+
np.array(range(self.length)) + np.random.normal(size=(self.length)),
|
|
27
|
+
np.array(range(self.length)) + np.random.normal(size=(self.length)),
|
|
31
28
|
]
|
|
32
29
|
)
|
|
33
30
|
|
|
@@ -49,8 +46,7 @@ class DummyLabelDataset(VisionDataset):
|
|
|
49
46
|
np.random.seed(12)
|
|
50
47
|
A = np.array(range(self.dim * self.dim)).reshape((self.dim, self.dim))
|
|
51
48
|
self.images = np.zeros(
|
|
52
|
-
shape=(self.length, self.n_channels,
|
|
53
|
-
self.dim, self.dim), dtype=np.float32
|
|
49
|
+
shape=(self.length, self.n_channels, self.dim, self.dim), dtype=np.float32
|
|
54
50
|
)
|
|
55
51
|
for idx in range(self.length):
|
|
56
52
|
for c in range(self.n_channels):
|
iapytoo/metrics/__init__.py
CHANGED
|
@@ -1,82 +0,0 @@
|
|
|
1
|
-
import torch
|
|
2
|
-
|
|
3
|
-
from iapytoo.utils.meta_singleton import MetaSingleton
|
|
4
|
-
|
|
5
|
-
from .predefined import R2Metric, RMSMetric, MSMetric, AccuracyMetric
|
|
6
|
-
from .metric import Metric
|
|
7
|
-
|
|
8
|
-
|
|
9
|
-
class MetricsCollection(Metric):
|
|
10
|
-
def __init__(self, tag: str, metric_names: list, config):
|
|
11
|
-
super().__init__(tag, config)
|
|
12
|
-
self.metrics = {}
|
|
13
|
-
factory = MetricFactory()
|
|
14
|
-
try:
|
|
15
|
-
for n in metric_names:
|
|
16
|
-
self.metrics[f"{tag}_{n}"] = factory.create_metric(n, config)
|
|
17
|
-
except MetricError as er:
|
|
18
|
-
print(f"Unable to create metric : {er}")
|
|
19
|
-
|
|
20
|
-
def to(self, device):
|
|
21
|
-
for m in self.metrics.values():
|
|
22
|
-
m.to(device)
|
|
23
|
-
|
|
24
|
-
# overwrite
|
|
25
|
-
def update(self, Y_hat, Y):
|
|
26
|
-
for m in self.metrics.values():
|
|
27
|
-
m.update(Y_hat, Y)
|
|
28
|
-
|
|
29
|
-
# overwrite
|
|
30
|
-
def compute(self):
|
|
31
|
-
self.results = {}
|
|
32
|
-
for k, m in self.metrics.items():
|
|
33
|
-
results = m.compute()
|
|
34
|
-
for name, result in results.items():
|
|
35
|
-
# prepend the name of the collection to all inner results : validation_accuracy for example
|
|
36
|
-
self.results[f'{self.name}_{name}'] = result
|
|
37
|
-
return self.results
|
|
38
|
-
|
|
39
|
-
# overwrite
|
|
40
|
-
def reset(self) -> None:
|
|
41
|
-
for m in self.metrics.values():
|
|
42
|
-
m.reset()
|
|
43
|
-
|
|
44
|
-
|
|
45
|
-
class MetricError(Exception):
|
|
46
|
-
def __init__(self, *args: object) -> None:
|
|
47
|
-
super().__init__(*args)
|
|
48
|
-
|
|
49
|
-
|
|
50
|
-
class MetricFactory(metaclass=MetaSingleton):
|
|
51
|
-
def __init__(self) -> None:
|
|
52
|
-
self.metrics_dict = {
|
|
53
|
-
"r2": R2Metric,
|
|
54
|
-
"rms": RMSMetric,
|
|
55
|
-
"ms": MSMetric,
|
|
56
|
-
"accuracy": AccuracyMetric,
|
|
57
|
-
}
|
|
58
|
-
|
|
59
|
-
def register_metric(self, key, metric_cls):
|
|
60
|
-
self.metrics_dict[key] = metric_cls
|
|
61
|
-
|
|
62
|
-
def create_metric(self, kind: str, config: dict, device="cpu") -> Metric:
|
|
63
|
-
"""Creates an architecture of NN
|
|
64
|
-
|
|
65
|
-
Args:
|
|
66
|
-
kind (str): kind of metric, key for the factory
|
|
67
|
-
config (dict): config dict to use to initialize metric
|
|
68
|
-
device (str, optional): Defaults to "cpu".
|
|
69
|
-
|
|
70
|
-
Raises:
|
|
71
|
-
MetricError: error raised if no architecture fit kind key
|
|
72
|
-
|
|
73
|
-
Returns:
|
|
74
|
-
iapytoo.metrics.Metric: metric
|
|
75
|
-
"""
|
|
76
|
-
try:
|
|
77
|
-
metric: Metric = self.metrics_dict[kind](config)
|
|
78
|
-
except KeyError:
|
|
79
|
-
raise MetricError(f"metric {kind} is not handled")
|
|
80
|
-
|
|
81
|
-
metric = metric.to(device=device)
|
|
82
|
-
return metric
|
|
@@ -0,0 +1,39 @@
|
|
|
1
|
+
from iapytoo.utils.config import Config
|
|
2
|
+
from iapytoo.metrics.metric import Metric
|
|
3
|
+
|
|
4
|
+
|
|
5
|
+
class MetricsCollection(Metric):
|
|
6
|
+
def __init__(self, tag: str, metric_names: list[str], config: Config):
|
|
7
|
+
from iapytoo.train.factories import Factory, MetricError
|
|
8
|
+
super().__init__(tag, config)
|
|
9
|
+
self.metrics = {}
|
|
10
|
+
try:
|
|
11
|
+
factory = Factory()
|
|
12
|
+
for n in metric_names:
|
|
13
|
+
self.metrics[f"{tag}_{n}"] = factory.create_metric(n, config)
|
|
14
|
+
except MetricError as er:
|
|
15
|
+
print(f"Unable to create metric : {er}")
|
|
16
|
+
|
|
17
|
+
def to(self, device):
|
|
18
|
+
for m in self.metrics.values():
|
|
19
|
+
m.to(device)
|
|
20
|
+
|
|
21
|
+
# overwrite
|
|
22
|
+
def update(self, Y_hat, Y):
|
|
23
|
+
for m in self.metrics.values():
|
|
24
|
+
m.update(Y_hat, Y)
|
|
25
|
+
|
|
26
|
+
# overwrite
|
|
27
|
+
def compute(self):
|
|
28
|
+
self.results = {}
|
|
29
|
+
for k, m in self.metrics.items():
|
|
30
|
+
results = m.compute()
|
|
31
|
+
for name, result in results.items():
|
|
32
|
+
# prepend the name of the collection to all inner results : validation_accuracy for example
|
|
33
|
+
self.results[f"{self.name}_{name}"] = result
|
|
34
|
+
return self.results
|
|
35
|
+
|
|
36
|
+
# overwrite
|
|
37
|
+
def reset(self) -> None:
|
|
38
|
+
for m in self.metrics.values():
|
|
39
|
+
m.reset()
|
iapytoo/metrics/metric.py
CHANGED
|
@@ -1,39 +1,46 @@
|
|
|
1
1
|
import torch
|
|
2
2
|
|
|
3
|
+
from iapytoo.utils.config import Config
|
|
4
|
+
|
|
3
5
|
|
|
4
6
|
class Metric:
|
|
5
|
-
def __init__(self, name, config, with_target=True):
|
|
7
|
+
def __init__(self, name, config: Config, with_target=True):
|
|
8
|
+
from iapytoo.train.factories import Factory
|
|
6
9
|
self.with_target = with_target
|
|
7
10
|
self.name = name
|
|
8
11
|
self.config = config
|
|
12
|
+
self.predictor = Factory().create_predictor(config)
|
|
9
13
|
|
|
10
14
|
# First dimension is for the concatenation.
|
|
11
|
-
self.
|
|
15
|
+
self.outputs = torch.zeros(size=(0,))
|
|
12
16
|
if self.with_target:
|
|
13
17
|
self.target = torch.zeros(size=(0,))
|
|
14
18
|
self.results = {}
|
|
15
19
|
|
|
16
20
|
@property
|
|
17
21
|
def device(self):
|
|
18
|
-
return self.
|
|
22
|
+
return self.outputs.device
|
|
23
|
+
|
|
24
|
+
@property
|
|
25
|
+
def predicted(self):
|
|
26
|
+
return self.predictor(self.outputs)
|
|
19
27
|
|
|
20
28
|
def to(self, device):
|
|
21
|
-
self.
|
|
29
|
+
self.outputs = self.outputs.to(device)
|
|
22
30
|
if self.with_target:
|
|
23
31
|
self.target = self.target.to(device)
|
|
24
32
|
return self
|
|
25
33
|
|
|
26
|
-
def update(self,
|
|
27
|
-
self.
|
|
34
|
+
def update(self, outputs, Y):
|
|
35
|
+
self.outputs = torch.cat((self.outputs, outputs), dim=0)
|
|
28
36
|
if self.with_target:
|
|
29
37
|
self.target = torch.cat((self.target, Y), dim=0)
|
|
30
38
|
|
|
31
39
|
def reset(self):
|
|
32
40
|
device = self.device
|
|
33
|
-
self.
|
|
41
|
+
self.outputs = torch.zeros(size=(0,), device=device)
|
|
34
42
|
if self.with_target:
|
|
35
|
-
self.target = torch.zeros(
|
|
36
|
-
size=self.output_size, device=self.device)
|
|
43
|
+
self.target = torch.zeros(size=(0,), device=self.device)
|
|
37
44
|
|
|
38
45
|
def compute(self):
|
|
39
46
|
pass
|
iapytoo/metrics/predefined.py
CHANGED
|
@@ -1,34 +1,35 @@
|
|
|
1
|
-
|
|
2
1
|
import torch
|
|
2
|
+
|
|
3
|
+
from iapytoo.utils.config import Config
|
|
3
4
|
from iapytoo.metrics.metric import Metric
|
|
4
5
|
|
|
5
6
|
|
|
6
7
|
class AccuracyMetric(Metric):
|
|
7
|
-
def __init__(self, config):
|
|
8
|
+
def __init__(self, config: Config):
|
|
8
9
|
super(AccuracyMetric, self).__init__("accuracy", config)
|
|
9
|
-
self.k = config
|
|
10
|
+
self.k = config.metrics.top_accuracy
|
|
10
11
|
|
|
11
12
|
def compute(self):
|
|
12
13
|
|
|
13
14
|
# Take care, for this metrics predicted and target do not have the same shape.
|
|
14
15
|
# predicted : ouputs of the models - kind of probabilities
|
|
15
16
|
# Target : label of the class
|
|
16
|
-
_, top_pred = self.
|
|
17
|
+
_, top_pred = self.outputs.topk(self.k, 1)
|
|
17
18
|
top_pred = top_pred.t()
|
|
18
19
|
correct = top_pred.eq(self.target.view(1, -1).expand_as(top_pred))
|
|
19
20
|
correct_1 = correct[:1].reshape(-1).float().sum(0, keepdim=True)
|
|
20
|
-
correct_k = correct[:self.k].reshape(-1).float().sum(0, keepdim=True)
|
|
21
|
+
correct_k = correct[: self.k].reshape(-1).float().sum(0, keepdim=True)
|
|
21
22
|
|
|
22
23
|
return {
|
|
23
|
-
|
|
24
|
-
|
|
25
|
-
|
|
26
|
-
|
|
24
|
+
"top-1": torch.round(100.0 * correct_1 / self.target.shape[0], decimals=2),
|
|
25
|
+
f"top-{self.k}": torch.round(
|
|
26
|
+
100.0 * correct_k / self.target.shape[0], decimals=2
|
|
27
|
+
),
|
|
27
28
|
}
|
|
28
29
|
|
|
29
30
|
|
|
30
31
|
class R2Metric(Metric):
|
|
31
|
-
def __init__(self, config) -> None:
|
|
32
|
+
def __init__(self, config: Config) -> None:
|
|
32
33
|
super(R2Metric, self).__init__("r2", config)
|
|
33
34
|
|
|
34
35
|
def compute(self):
|
|
@@ -49,7 +50,7 @@ class R2Metric(Metric):
|
|
|
49
50
|
|
|
50
51
|
|
|
51
52
|
class MSMetric(Metric):
|
|
52
|
-
def __init__(self, config) -> None:
|
|
53
|
+
def __init__(self, config: Config) -> None:
|
|
53
54
|
super(MSMetric, self).__init__("mean_square", config)
|
|
54
55
|
|
|
55
56
|
def _compute(self):
|
|
@@ -63,7 +64,7 @@ class MSMetric(Metric):
|
|
|
63
64
|
|
|
64
65
|
|
|
65
66
|
class RMSMetric(MSMetric):
|
|
66
|
-
def __init__(self, config) -> None:
|
|
67
|
+
def __init__(self, config: Config) -> None:
|
|
67
68
|
Metric.__init__(self, "root_mean_square", config)
|
|
68
69
|
self.name = "rms"
|
|
69
70
|
|
iapytoo/predictions/__init__.py
CHANGED
|
@@ -1,20 +1,35 @@
|
|
|
1
|
+
from __future__ import annotations
|
|
1
2
|
import torch
|
|
2
|
-
import numpy as np
|
|
3
3
|
|
|
4
4
|
from .types import PredictionType
|
|
5
5
|
from .plotters import CollectionPlotters
|
|
6
|
-
|
|
6
|
+
|
|
7
|
+
from typing import TYPE_CHECKING
|
|
8
|
+
|
|
9
|
+
if TYPE_CHECKING:
|
|
10
|
+
from iapytoo.train.inference import Inference
|
|
7
11
|
|
|
8
12
|
|
|
9
13
|
class Predictions:
|
|
10
|
-
def __init__(self,
|
|
11
|
-
self.norm = norm
|
|
14
|
+
def __init__(self, inference: Inference):
|
|
12
15
|
self.outputs = None # dimension is assigned in compute
|
|
13
16
|
self.actual = None
|
|
14
|
-
|
|
17
|
+
|
|
18
|
+
self.inference = inference
|
|
19
|
+
|
|
15
20
|
self.prediction_plotter: CollectionPlotters = CollectionPlotters()
|
|
16
21
|
self.prediction_plotter.connect(self)
|
|
17
22
|
|
|
23
|
+
@property
|
|
24
|
+
def valuator(self):
|
|
25
|
+
assert self.inference is not None, "no inference for this prediction"
|
|
26
|
+
return self.inference.valuator
|
|
27
|
+
|
|
28
|
+
@property
|
|
29
|
+
def predictor(self):
|
|
30
|
+
assert self.inference is not None, "no inference for this prediction"
|
|
31
|
+
return self.inference.predictor
|
|
32
|
+
|
|
18
33
|
def __len__(self):
|
|
19
34
|
"""return the number of plotters used by the predictions"""
|
|
20
35
|
return len(self.prediction_plotter)
|
|
@@ -22,35 +37,27 @@ class Predictions:
|
|
|
22
37
|
def add_plotter(self, prediction_plotter):
|
|
23
38
|
self.prediction_plotter.add(prediction_plotter)
|
|
24
39
|
|
|
25
|
-
def compute(self,
|
|
26
|
-
|
|
27
|
-
|
|
28
|
-
self.
|
|
29
|
-
self.actual = torch.zeros(size=(0,), device=device)
|
|
40
|
+
def compute(self, loader):
|
|
41
|
+
|
|
42
|
+
assert self.valuator is not None, "no valuator"
|
|
43
|
+
assert self.predictor is not None, "no predictor"
|
|
30
44
|
|
|
31
|
-
|
|
32
|
-
|
|
33
|
-
with torch.no_grad():
|
|
34
|
-
for X, Y in loader:
|
|
35
|
-
X = X.to(device)
|
|
36
|
-
Y = Y.to(device)
|
|
37
|
-
model_output = model(X)
|
|
45
|
+
self.outputs = torch.zeros(size=(0,))
|
|
46
|
+
self.actual = torch.zeros(size=(0,))
|
|
38
47
|
|
|
39
|
-
|
|
40
|
-
|
|
48
|
+
for outputs, actual in self.valuator.evaluate_loader(loader):
|
|
49
|
+
self.outputs = torch.cat((self.outputs, outputs), dim=0)
|
|
41
50
|
|
|
42
|
-
|
|
43
|
-
|
|
44
|
-
self.actual = y_scaling.inv(self.actual)
|
|
51
|
+
if actual is not None:
|
|
52
|
+
self.actual = torch.cat((self.actual, actual), dim=0)
|
|
45
53
|
|
|
46
54
|
def tensor(self, type: PredictionType = PredictionType.PREDICTED):
|
|
47
55
|
if type == PredictionType.PREDICTED:
|
|
48
|
-
|
|
49
|
-
return predicted.detach().cpu()
|
|
56
|
+
return self.predictor(self.outputs)
|
|
50
57
|
elif type == PredictionType.ACTUAL:
|
|
51
|
-
return self.actual
|
|
58
|
+
return self.actual
|
|
52
59
|
else:
|
|
53
|
-
return self.outputs
|
|
60
|
+
return self.outputs
|
|
54
61
|
|
|
55
62
|
def numpy(self, type: PredictionType = PredictionType.PREDICTED):
|
|
56
63
|
return self.tensor(type).numpy()
|
|
@@ -60,19 +67,3 @@ class Predictions:
|
|
|
60
67
|
return self.prediction_plotter.plot(epoch)
|
|
61
68
|
else:
|
|
62
69
|
return {} # nothing to plot
|
|
63
|
-
|
|
64
|
-
|
|
65
|
-
class GenerativePredictions(Predictions):
|
|
66
|
-
def __init__(self, norm=False):
|
|
67
|
-
super().__init__(norm)
|
|
68
|
-
|
|
69
|
-
def compute(self, training, loader):
|
|
70
|
-
device = training.device
|
|
71
|
-
self.outputs = torch.zeros(size=(0,), device=device)
|
|
72
|
-
generator = training.generator
|
|
73
|
-
generator.eval()
|
|
74
|
-
with torch.no_grad():
|
|
75
|
-
for X in loader:
|
|
76
|
-
X = X.to(device)
|
|
77
|
-
gen_output = generator(X)
|
|
78
|
-
self.outputs = torch.cat((self.outputs, gen_output), dim=0)
|
iapytoo/predictions/feature.py
CHANGED
|
@@ -11,6 +11,5 @@ class Feature:
|
|
|
11
11
|
for i in range(predictions.actual.shape[0]):
|
|
12
12
|
actual_del = self.computer(predictions.actual[i, 0, :], **kwargs)
|
|
13
13
|
self.actual.append(actual_del)
|
|
14
|
-
predicted_del = self.computer(
|
|
15
|
-
predictions.outputs[i, 0, :], **kwargs)
|
|
14
|
+
predicted_del = self.computer(predictions.outputs[i, 0, :], **kwargs)
|
|
16
15
|
self.predicted.append(predicted_del)
|
iapytoo/predictions/plotters.py
CHANGED
|
@@ -7,10 +7,12 @@ from scipy.signal import welch
|
|
|
7
7
|
import numpy as np
|
|
8
8
|
|
|
9
9
|
from iapytoo.predictions.types import PredictionType
|
|
10
|
+
from iapytoo.utils.config import Config
|
|
11
|
+
from iapytoo.utils.singleton import singleton
|
|
10
12
|
|
|
11
13
|
|
|
12
14
|
class PredictionPlotter:
|
|
13
|
-
def __init__(self, title=None):
|
|
15
|
+
def __init__(self, title=None, config: Config | None = None):
|
|
14
16
|
self.predictions = None
|
|
15
17
|
self.title = title
|
|
16
18
|
|
|
@@ -23,7 +25,7 @@ class PredictionPlotter:
|
|
|
23
25
|
|
|
24
26
|
class CollectionPlotters(PredictionPlotter):
|
|
25
27
|
|
|
26
|
-
def __init__(self, title=None):
|
|
28
|
+
def __init__(self, title=None, config: Config | None = None):
|
|
27
29
|
super(CollectionPlotters, self).__init__(title)
|
|
28
30
|
self.plotters = []
|
|
29
31
|
|
|
@@ -43,7 +45,7 @@ class CollectionPlotters(PredictionPlotter):
|
|
|
43
45
|
|
|
44
46
|
|
|
45
47
|
class ScatterPlotter(PredictionPlotter):
|
|
46
|
-
def __init__(self):
|
|
48
|
+
def __init__(self, config: Config | None = None):
|
|
47
49
|
super().__init__(title="actual_versus_predicted")
|
|
48
50
|
|
|
49
51
|
def plot(self, epoch):
|
|
@@ -56,7 +58,7 @@ class ScatterPlotter(PredictionPlotter):
|
|
|
56
58
|
|
|
57
59
|
|
|
58
60
|
class ConfusionPlotter(PredictionPlotter):
|
|
59
|
-
def __init__(self):
|
|
61
|
+
def __init__(self, config: Config | None = None):
|
|
60
62
|
super().__init__(title="confusion_matrix")
|
|
61
63
|
|
|
62
64
|
def plot(self, epoch):
|
|
@@ -72,7 +74,7 @@ class ConfusionPlotter(PredictionPlotter):
|
|
|
72
74
|
|
|
73
75
|
class TSNEPlotter(PredictionPlotter):
|
|
74
76
|
|
|
75
|
-
def __init__(self):
|
|
77
|
+
def __init__(self, config: Config | None = None):
|
|
76
78
|
super().__init__(title="tsne")
|
|
77
79
|
|
|
78
80
|
def plot(self, epoch):
|
|
@@ -98,7 +100,7 @@ class TSNEPlotter(PredictionPlotter):
|
|
|
98
100
|
|
|
99
101
|
|
|
100
102
|
class Fake1DPlotter(PredictionPlotter):
|
|
101
|
-
def __init__(self, n_plot=4):
|
|
103
|
+
def __init__(self, n_plot=4, config: Config | None = None):
|
|
102
104
|
super().__init__(title="generated_fake1d")
|
|
103
105
|
self.n_plot = n_plot
|
|
104
106
|
|
|
@@ -107,7 +109,7 @@ class Fake1DPlotter(PredictionPlotter):
|
|
|
107
109
|
f, a = plt.subplots(self.n_plot, self.n_plot, figsize=(8, 8))
|
|
108
110
|
for i in range(self.n_plot):
|
|
109
111
|
for j in range(self.n_plot):
|
|
110
|
-
a[i][j].plot(fake[i * self.n_plot + j]
|
|
112
|
+
a[i][j].plot(fake[i * self.n_plot + j, 0, :])
|
|
111
113
|
a[i][j].set_xticks(())
|
|
112
114
|
a[i][j].set_yticks(())
|
|
113
115
|
|
|
@@ -115,23 +117,26 @@ class Fake1DPlotter(PredictionPlotter):
|
|
|
115
117
|
|
|
116
118
|
|
|
117
119
|
class DSPPlotter(PredictionPlotter):
|
|
118
|
-
def __init__(self, f_max=4, nperseg=200):
|
|
120
|
+
def __init__(self, f_max=4, nperseg=200, config: Config | None = None):
|
|
119
121
|
super().__init__(title="dsp")
|
|
120
122
|
self.f_max = f_max
|
|
121
123
|
self.nperseg = nperseg
|
|
122
124
|
|
|
123
125
|
def plot(self, epoch):
|
|
124
|
-
fake = self.predictions.
|
|
126
|
+
fake = self.predictions.numpy(PredictionType.PREDICTED)
|
|
127
|
+
assert fake is not None and fake.shape[0] > 0, "no signal for DSP extraction"
|
|
125
128
|
f, ax = plt.subplots()
|
|
126
129
|
|
|
127
|
-
frequencies, power_spectrum = welch(
|
|
130
|
+
frequencies, power_spectrum = welch(
|
|
131
|
+
fake[0, 0, :], fs=self.f_max, nperseg=self.nperseg
|
|
132
|
+
)
|
|
128
133
|
ax.plot(frequencies, power_spectrum)
|
|
129
134
|
|
|
130
135
|
return {self.title: f}
|
|
131
136
|
|
|
132
137
|
|
|
133
138
|
class Fake2DPlotter(PredictionPlotter):
|
|
134
|
-
def __init__(self, n_plot=4):
|
|
139
|
+
def __init__(self, n_plot=4, config: Config | None = None):
|
|
135
140
|
super().__init__("generated_fake2d")
|
|
136
141
|
self.n_plot = n_plot
|
|
137
142
|
|
|
@@ -144,3 +149,26 @@ class Fake2DPlotter(PredictionPlotter):
|
|
|
144
149
|
ax.imshow(grid_img.permute(1, 2, 0))
|
|
145
150
|
|
|
146
151
|
return {self.title: f}
|
|
152
|
+
|
|
153
|
+
|
|
154
|
+
@singleton
|
|
155
|
+
class PlotterFactory:
|
|
156
|
+
def __init__(self) -> None:
|
|
157
|
+
self.plotter_dict: dict[str, type[PredictionPlotter]] = {
|
|
158
|
+
"scatter": ScatterPlotter,
|
|
159
|
+
"confusion": ConfusionPlotter,
|
|
160
|
+
"tsne": TSNEPlotter,
|
|
161
|
+
"fake1d": Fake1DPlotter,
|
|
162
|
+
"fake2d": Fake2DPlotter,
|
|
163
|
+
"dsp": DSPPlotter,
|
|
164
|
+
}
|
|
165
|
+
|
|
166
|
+
def register_plotter(self, key: str, plotter_cls: type[PredictionPlotter]) -> None:
|
|
167
|
+
self.plotter_dict[key] = plotter_cls
|
|
168
|
+
|
|
169
|
+
def create_plotter(self, kind, **kwargs) -> PredictionPlotter:
|
|
170
|
+
try:
|
|
171
|
+
plotter: PredictionPlotter = self.plotter_dict[kind](**kwargs)
|
|
172
|
+
except KeyError:
|
|
173
|
+
raise KeyError(f"Plotter {kind} doesn't exist")
|
|
174
|
+
return plotter
|
iapytoo/train/checkpoint.py
CHANGED
|
@@ -7,22 +7,22 @@ import tempfile
|
|
|
7
7
|
class CheckPoint:
|
|
8
8
|
def __init__(self, run_id=None) -> None:
|
|
9
9
|
self.params = {"run_id": run_id}
|
|
10
|
-
|
|
11
|
-
|
|
12
|
-
assert run is not None, f"unable to find run {self.run_id}"
|
|
10
|
+
artifact_uri = self.artifact_uri
|
|
11
|
+
if artifact_uri is not None:
|
|
13
12
|
with tempfile.TemporaryDirectory() as tmpdir:
|
|
14
|
-
|
|
15
|
-
|
|
13
|
+
ck_path = mlflow.artifacts.download_artifacts(
|
|
14
|
+
artifact_uri=artifact_uri,
|
|
15
|
+
dst_path=tmpdir
|
|
16
16
|
)
|
|
17
|
-
ck_path = os.path.join(local_artifact, "checkpoints")
|
|
18
17
|
files = []
|
|
19
18
|
for f in os.listdir(ck_path):
|
|
20
19
|
ck_name = os.path.join(ck_path, f)
|
|
21
|
-
if os.path.isfile(ck_name):
|
|
20
|
+
if "checkpoint" in f and os.path.isfile(ck_name):
|
|
22
21
|
files.append(ck_name)
|
|
23
22
|
files.sort(key=lambda x: os.path.getmtime(x))
|
|
24
23
|
ck_name = files[-1]
|
|
25
|
-
self.params = torch.load(
|
|
24
|
+
self.params = torch.load(
|
|
25
|
+
ck_name, weights_only=False)
|
|
26
26
|
else:
|
|
27
27
|
self.params["epoch"] = -1
|
|
28
28
|
|
|
@@ -30,6 +30,14 @@ class CheckPoint:
|
|
|
30
30
|
def run_id(self):
|
|
31
31
|
return self.params["run_id"]
|
|
32
32
|
|
|
33
|
+
@property
|
|
34
|
+
def artifact_uri(self):
|
|
35
|
+
if self.run_id is None:
|
|
36
|
+
return None
|
|
37
|
+
run = mlflow.get_run(self.run_id)
|
|
38
|
+
assert run is not None, f"unable to find run {self.run_id}"
|
|
39
|
+
return os.path.join(run.info.artifact_uri, "checkpoints")
|
|
40
|
+
|
|
33
41
|
@property
|
|
34
42
|
def epoch(self):
|
|
35
43
|
return self.params["epoch"]
|
|
@@ -41,4 +49,5 @@ class CheckPoint:
|
|
|
41
49
|
|
|
42
50
|
def init(self, training, only_model=False):
|
|
43
51
|
if self.run_id is not None:
|
|
44
|
-
training.load_state_dict(
|
|
52
|
+
training.load_state_dict(
|
|
53
|
+
self.params["training"], only_model=only_model)
|