MNT remove seaborn as a dependency (#21)

Co-authored-by: mathurinm <[email protected]>

Author: Badr-MOUFAD

.gitignore

@@ -1,7 +1,11 @@
-/skglm.egg-info/*
+*.egg-info
 *.pyc
 
 /doc/_build/
 /doc/gen_modules/
 /doc/generated/
 /doc/auto_examples/
+
+__pycache__
+__cache__
+*_cache

doc/doc-requirements.txt

@@ -1,5 +1,8 @@
 benchopt
+libsvmdata>=0.2
+matplotlib>=2.0.0
 numpydoc
 pillow
 sphinx-bootstrap-theme
-sphinx-gallery
+sphinx-gallery
+pytest

examples/plot_sparse_recovery.py

@@ -10,7 +10,6 @@
 #         Quentin Klopfenstein
 
 import numpy as np
-import seaborn as sns
 from numpy.linalg import norm
 import matplotlib.pyplot as plt
 from sklearn.model_selection import train_test_split
@@ -21,7 +20,7 @@
 from skglm.datafits import Quadratic
 from skglm.penalties import L1, MCPenalty, L0_5, L2_3
 
-current_palette = sns.color_palette("colorblind")
+cmap = plt.get_cmap('tab10')
 
 # Simulate sparse data
 n_features = 1000
@@ -56,10 +55,10 @@
 penalties['l23'] = L2_3(alpha=1)
 
 colors = {}
-colors['lasso'] = current_palette[0]
-colors['mcp'] = current_palette[1]
-colors['l05'] = current_palette[2]
-colors['l23'] = current_palette[3]
+colors['lasso'] = cmap(0)
+colors['mcp'] = cmap(1)
+colors['l05'] = cmap(2)
+colors['l23'] = cmap(3)
 
 f1 = {}
 estimation_error = {}

setup.py

@@ -12,7 +12,9 @@
     raise RuntimeError('Could not determine version')
 
 DISTNAME = 'skglm'
-DESCRIPTION = 'Fast and modular scikit-learn replacement for generalized linear models'
+DESCRIPTION = 'A fast and modular scikit-learn replacement for generalized linear models'
+with open('README.rst', 'r') as f:
+    LONG_DESCRIPTION = f.read()
 MAINTAINER = 'Mathurin Massias'
 MAINTAINER_EMAIL = '[email protected]'
 LICENSE = 'BSD (3-clause)'
@@ -23,13 +25,12 @@
 setup(name=DISTNAME,
       version=version,
       description=DESCRIPTION,
-      long_description=open('README.rst').read(),
+      long_description=LONG_DESCRIPTION,
       maintainer=MAINTAINER,
       maintainer_email=MAINTAINER_EMAIL,
       url=URL,
       download_url=DOWNLOAD_URL,
       packages=find_packages(),
-      install_requires=['libsvmdata>=0.2', 'numpy>=1.12', 'numba',
-                        'seaborn>=0.7',
-                        'joblib', 'scipy>=0.18.0', 'matplotlib>=2.0.0',
-                        'scikit-learn>=1.0', 'pandas', 'ipython'],)
+      install_requires=['numpy>=1.12', 'numba',
+                        'scipy>=0.18.0', 'scikit-learn>=1.0']
+      )

skglm/utils.py

@@ -1,14 +1,10 @@
 import numpy as np
-import seaborn as sns
 from numba import njit
 
 from numpy.linalg import norm
 from sklearn.utils import check_random_state
 
 
-C_LIST = sns.color_palette()
-
-
 @njit
 def ST(x, u):
     """Soft-thresholding of scalar x at level u."""