New split volume program #768
Conversation
|
![github-advanced-security[bot]](https://avatars.githubusercontent.com/in/57789?s=60&v=4){kind=small}
| averages = self._computeAverages(data, gains, noise2) | ||
|
|
||
| if nComponents == 0: | ||
| pca = sklearn.decomposition.PCA(n_components='mle') |
Check notice
Code scanning / SonarCloud
Results that depend on random number generation should be reproducible Low
| if nComponents == 0: | ||
| pca = sklearn.decomposition.PCA(n_components='mle') | ||
| else: | ||
| pca = sklearn.decomposition.PCA(n_components=nComponents) |
Check notice
Code scanning / SonarCloud
Results that depend on random number generation should be reproducible Low
| if method == 'isomap': | ||
| result = sklearn.manifold.Isomap(n_components=d, n_neighbors=neighbors, n_jobs=jobs) | ||
| elif method == 'spectral': | ||
| result = sklearn.manifold.SpectralEmbedding(n_components=d, n_neighbors=neighbors, n_jobs=jobs) |
Check notice
Code scanning / SonarCloud
Results that depend on random number generation should be reproducible Low
| elif method == 'spectral': | ||
| result = sklearn.manifold.SpectralEmbedding(n_components=d, n_neighbors=neighbors, n_jobs=jobs) | ||
| elif method == 'tsne': | ||
| result = sklearn.manifold.TSNE(n_components=d, n_jobs=jobs) |
Check notice
Code scanning / SonarCloud
Results that depend on random number generation should be reproducible Low
| elif method == 'tsne': | ||
| result = sklearn.manifold.TSNE(n_components=d, n_jobs=jobs) | ||
| elif method == 'mds': | ||
| result = sklearn.manifold.MDS(n_components=d, n_jobs=jobs) |
Check notice
Code scanning / SonarCloud
Results that depend on random number generation should be reproducible Low
| covarianceType = self.COVARIANCE_TYPES[self.covarianceType.get()] | ||
| components = self.classCount.get() | ||
|
|
||
| return sklearn.mixture.GaussianMixture( |
Check notice
Code scanning / SonarCloud
Results that depend on random number generation should be reproducible Low
Complete overhaul of the split volume program. A new graph theory-based approach was implemented to perform 3D classification