#include <vector>
#include "NumCpp.hpp"
Go to the source code of this file.
|
| nc::NdArray< float > | PCA::scale (const nc::NdArray< float > &sample) |
| | Scales the features using a pre-built StandardScaler. More...
|
| |
| nc::NdArray< float > | PCA::scale (const nc::NdArray< float > &samples, nc::NdArray< float > &extMean, nc::NdArray< float > &extScale) |
| | Scales the features using a pre-built StandardScaler. This version is for a batch of samples. More...
|
| |
| nc::NdArray< float > | PCA::transform (const nc::NdArray< float > &sample) |
| | Transforms the sample using a PCA model previously built. More...
|
| |
| nc::NdArray< float > | PCA::transform (const nc::NdArray< float > &samples, const nc::NdArray< float > &extMean) |
| | Transforms the sample using a PCA model previously built. This version is for a batch of samples. More...
|
| |
| nc::NdArray< float > | PCA::PCA (const std::vector< float > &sample) |
| | Applies the pipeline Standard Scaler plus PCA to the sample data. More...
|
| |
| nc::NdArray< float > | PCA::PCA (const nc::NdArray< float > &sample) |
| | Applies the pipeline Standard Scaler plus PCA to the sample data. More...
|
| |
| nc::NdArray< float > | PCA::PCABatch (const nc::NdArray< float > &samples) |
| | Applies the pipeline Standard Scaler plus PCA to the batch of samples. More...
|
| |
|
| nc::NdArray< float > | PCA::componentsT |
| | This is the PCA components.T we obtained when using 5000 samples and considering only the 8-dimensional features space. HERE WE DO NOT USE THE INSTANCE DATA TO CREATE THE PCA and we ensure a balanced distribution of the targets. More...
|
| |
| nc::NdArray< float > | PCA::mean |
| | This is the PCA mean we obtained using only 5000 samples and considering only the 8-dimensional features space. HERE WE DO NOT USE THE INSTANCE DATA TO CREATE THE PCA and we ensure a balanced distribution of the targets. More...
|
| |
| nc::NdArray< float > | PCA::scaler_mean |
| | This is the scaler mean we obtained using only 5000 samples and considering only the 8-dimensional features space. HERE WE DO NOT USE THE INSTANCE DATA TO CREATE THE PCA. More...
|
| |
| nc::NdArray< float > | PCA::scaler_scale |
| | This is the scaler scale obtained with only 5000 samples and considering only the 8-dimensional features space. HERE WE DO NOT USE THE INSTANCE DATA TO CREATE THE PCA. More...
|
| |
- Author
- Alejandro Marrero (amarr.nosp@m.erd@.nosp@m.ull.e.nosp@m.du.e.nosp@m.s)
- Version
- 0.1
- Date
- 2022-10-21
- Copyright
- Copyright (c) 2022
◆ PCA() [1/2]
| nc::NdArray<float> PCA::PCA |
( |
const nc::NdArray< float > & |
sample | ) |
|
Applies the pipeline Standard Scaler plus PCA to the sample data.
- Parameters
-
- Returns
- nc::NdArray<float>
◆ PCA() [2/2]
| nc::NdArray<float> PCA::PCA |
( |
const std::vector< float > & |
sample | ) |
|
Applies the pipeline Standard Scaler plus PCA to the sample data.
- Parameters
-
- Returns
- nc::NdArray<float>
◆ PCABatch()
| nc::NdArray<float> PCA::PCABatch |
( |
const nc::NdArray< float > & |
samples | ) |
|
Applies the pipeline Standard Scaler plus PCA to the batch of samples.
- Parameters
-
- Returns
- nc::NdArray<float>
◆ scale() [1/2]
| nc::NdArray<float> PCA::scale |
( |
const nc::NdArray< float > & |
sample | ) |
|
Scales the features using a pre-built StandardScaler.
- Parameters
-
- Returns
- nc::NdArray<float>
◆ scale() [2/2]
| nc::NdArray<float> PCA::scale |
( |
const nc::NdArray< float > & |
samples, |
|
|
nc::NdArray< float > & |
extMean, |
|
|
nc::NdArray< float > & |
extScale |
|
) |
| |
Scales the features using a pre-built StandardScaler. This version is for a batch of samples.
- Parameters
-
- Returns
- nc::NdArray<float>
◆ transform() [1/2]
| nc::NdArray<float> PCA::transform |
( |
const nc::NdArray< float > & |
sample | ) |
|
Transforms the sample using a PCA model previously built.
- Parameters
-
- Returns
- nc::NdArray<float>
◆ transform() [2/2]
| nc::NdArray<float> PCA::transform |
( |
const nc::NdArray< float > & |
samples, |
|
|
const nc::NdArray< float > & |
extMean |
|
) |
| |
Transforms the sample using a PCA model previously built. This version is for a batch of samples.
- Parameters
-
- Returns
- nc::NdArray<float>
◆ componentsT
| nc::NdArray<float> PCA::componentsT |
Initial value:= {
{7.375311811592301370e-02, -3.178655266058625917e-01},
{4.093087483174921992e-01, 4.955981249916805442e-01},
{3.346693276473038936e-01, -6.128877541751576175e-01},
{5.953059210206299579e-01, 1.328377255939875989e-01},
{-5.162302414308733711e-02, -4.926638878583582404e-01},
{5.983052657524839946e-01, -1.317157490269994513e-01},
}
This is the PCA components.T we obtained when using 5000 samples and considering only the 8-dimensional features space. HERE WE DO NOT USE THE INSTANCE DATA TO CREATE THE PCA and we ensure a balanced distribution of the targets.
◆ mean
| nc::NdArray<float> PCA::mean |
Initial value:= {
-1.302661682226850347e-16, 5.921189464667501929e-18,
-2.107943449421630714e-15, 9.577523959099684045e-16,
5.329070518200751197e-17, -3.907985046680551042e-16,
}
This is the PCA mean we obtained using only 5000 samples and considering only the 8-dimensional features space. HERE WE DO NOT USE THE INSTANCE DATA TO CREATE THE PCA and we ensure a balanced distribution of the targets.
◆ scaler_mean
| nc::NdArray<float> PCA::scaler_mean |
Initial value:= {
1.079515125000000080e+04, 2.652374999999999972e+01,
2.904959681383768952e+02, 5.040460461552937659e+02,
1.784837500024586854e+00, 9.787104166666666742e+02,
}
This is the scaler mean we obtained using only 5000 samples and considering only the 8-dimensional features space. HERE WE DO NOT USE THE INSTANCE DATA TO CREATE THE PCA.
◆ scaler_scale
| nc::NdArray<float> PCA::scaler_scale |
Initial value:= {
6.728206620826987091e+03, 2.899059219708179924e+01,
1.302756303387655201e+01, 2.544038827736384079e+01,
1.753588622119440776e+00, 1.909781553371280793e+01,
}
This is the scaler scale obtained with only 5000 samples and considering only the 8-dimensional features space. HERE WE DO NOT USE THE INSTANCE DATA TO CREATE THE PCA.
1.9.1