Graph model classes

.gitignore

@@ -2,7 +2,7 @@
 __pycache__/
 *.py[cod]
 *$py.class
-*.csv
+# *.csv
 
 # C extensions
 *.so
@@ -123,4 +123,4 @@ notebooks/bpedigo/diag_procrust.py
 notebooks/bpedigo/new_pairs.py
 *.code-workspace
 *.png
-
+notebooks/bpedigo

docs/reference/datasets.rst

@@ -0,0 +1,11 @@
+Datasets
+========
+
+.. currentmodule:: graspy.datasets
+
+Drosophila larval mushroom body
+-------------------------------
+
+.. autofunction:: load_drosophila_left
+
+.. autofunction:: load_drosophila_right

docs/reference/index.rst

@@ -7,9 +7,10 @@ Reference
  :maxdepth: 2
 
  simulations
+ models
  embed
  cluster
  inference
  plotting
- utils
-
+ datasets
+ utils

docs/reference/models.rst

@@ -0,0 +1,23 @@
+Models
+======
+
+.. currentmodule:: graspy.models
+
+Erdos-Reyni models
+------------------
+
+.. autoclass:: EREstimator
+
+.. autoclass:: DCEREstimator
+
+Stochastic block models
+-----------------------
+
+.. autoclass:: SBEstimator
+
+.. autoclass:: DCSBEstimator
+
+Latent position models
+----------------------
+
+.. autoclass:: RDPGEstimator

graspy/__init__.py

@@ -10,6 +10,7 @@
 import graspy.plot
 import graspy.simulations
 import graspy.utils
-
+import graspy.models
+import graspy.datasets
 
 __version__ = "0.0.2"

graspy/cluster/gclust.py

@@ -35,11 +35,12 @@ class GaussianCluster(BaseCluster):
  ----------
  min_components : int, defaults to 1. 
  The minimum number of mixture components to consider (unless
- max_components=None, in which case the parameter is one the highest
- number of mixture components to consider and must be less than
- max_components).
+ max_components=None, in which case this is the maximum number of
+ components to consider). If max_componens is not None, min_components
+ must be less than or equal to max_components.
  max_components : int, defaults to 1.
- The maximum number of mixture components to consider.
+ The maximum number of mixture components to consider. Must be greater 
+ than or equal to min_components.
 
  covariance_type : {'full' (default), 'tied', 'diag', 'spherical'}, optional
  String or list/array describing the type of covariance parameters to use.
@@ -83,7 +84,7 @@ class GaussianCluster(BaseCluster):
 
  def __init__(
  self,
- min_components,
+ min_components=2,
  max_components=None,
  covariance_type="full",
  random_state=None,
@@ -102,8 +103,8 @@ def __init__(
  if max_components <= 0:
  msg = "max_components must be >= 1 or None."
  raise ValueError(msg)
- elif min_components >= max_components:
- msg = "min_components must be strictly smaller than max_components."
+ elif min_components > max_components:
+ msg = "min_components must be less than or equal to max_components."
  raise ValueError(msg)
  elif max_components is not None:
  msg = "max_components must be an integer or None, not {}.".format(

graspy/datasets/__init__.py

@@ -0,0 +1,3 @@
+from .base import load_drosophila_left, load_drosophila_right
+
+__all__ = ["load_drosophila_left", "load_drosophila_right"]

graspy/datasets/base.py

@@ -0,0 +1,94 @@
+from os.path import dirname, join
+import numpy as np
+
+
+def load_drosophila_left(return_labels=False):
+ """
+ Load the left Drosophila larva mushroom body connectome
+
+ The mushroom body is a learning and memory center in the fly 
+ brain which is involved in sensory integration and processing. 
+ This connectome was observed by electron microscopy and then
+ individial neurons were reconstructed; synaptic partnerships
+ between these neurons became the edges of the graph. 
+
+ Parameters
+ ----------
+ return_labels : bool, optional (default=False)
+ whether to have a second return value which is an array of
+ cell type labels for each node in the adjacency matrix
+
+ Returns
+ -------
+ graph : np.ndarray
+ Adjacency matrix of the connectome
+ labels : np.ndarray
+ Only returned if `return_labels` is true. Array of
+ string labels for each cell (vertex)
+
+ References
+ ----------
+ .. [1] Eichler, K., Li, F., Litwin-Kumar, A., Park, Y., Andrade, I.,
+ Schneider-Mizell, C. M., ... & Fetter, R. D. (2017). The
+ complete connectome of a learning and memory centre in an insect
+ brain. Nature, 548(7666), 175.
+ """
+
+ module_path = dirname(__file__)
+ folder = "drosophila"
+ filename = "left_adjacency.csv"
+ with open(join(module_path, folder, filename)) as csv_file:
+ graph = np.loadtxt(csv_file, dtype=int)
+ if return_labels:
+ filename = "left_cell_labels.csv"
+ with open(join(module_path, folder, filename)) as csv_file:
+ labels = np.loadtxt(csv_file, dtype=str)
+ return graph, labels
+ else:
+ return graph
+
+
+def load_drosophila_right(return_labels=False):
+ """
+ Load the right Drosophila larva mushroom body connectome
+
+ The mushroom body is a learning and memory center in the fly 
+ brain which is involved in sensory integration and processing. 
+ This connectome was observed by electron microscopy and then
+ individial neurons were reconstructed; synaptic partnerships
+ between these neurons became the edges of the graph. 
+
+ Parameters
+ ----------
+ return_labels : bool, optional (default=False)
+ whether to have a second return value which is an array of
+ cell type labels for each node in the adjacency matrix
+
+ Returns
+ -------
+ graph : np.ndarray
+ Adjacency matrix of the connectome
+ labels : np.ndarray
+ Only returned if `return_labels` is true. Array of
+ string labels for each cell (vertex)
+
+ References
+ ----------
+ .. [1] Eichler, K., Li, F., Litwin-Kumar, A., Park, Y., Andrade, I.,
+ Schneider-Mizell, C. M., ... & Fetter, R. D. (2017). The
+ complete connectome of a learning and memory centre in an insect
+ brain. Nature, 548(7666), 175.
+ """
+
+ module_path = dirname(__file__)
+ folder = "drosophila"
+ filename = "right_adjacency.csv"
+ with open(join(module_path, folder, filename)) as csv_file:
+ graph = np.loadtxt(csv_file, dtype=int)
+ if return_labels:
+ filename = "right_cell_labels.csv"
+ with open(join(module_path, folder, filename)) as csv_file:
+ labels = np.loadtxt(csv_file, dtype=str)
+ return graph, labels
+ else:
+ return graph