creating new release

README.rst

@@ -5,6 +5,8 @@ UNCURL
     :target: https://travis-ci.org/yjzhang/uncurl_python
 .. image:: https://img.shields.io/pypi/v/uncurl-seq.svg
     :target: https://pypi.python.org/pypi/uncurl-seq
+.. image:: https://pepy.tech/badge/uncurl-seq
+    :target: https://pepy.tech/project/uncurl-seq
 
 
 Installation
@@ -16,7 +18,7 @@ Alternatively, uncurl can be installed from source: After cloning the repository
 
 Requirements: numpy, scipy, cython, scikit-learn
 
-Tested on python 2.7, 3.4 on Linux.
+Tested on python 2.7, 3.5 on Linux.
 
 For parallel state estimation, OpenMP is required.
 

setup.py

@@ -24,7 +24,7 @@
     long_description = f.read()
 
 setup(name='uncurl_seq',
-      version='0.2.11',
+      version='0.2.12',
       description='Tool for pre-processing single-cell RNASeq data',
       long_description=long_description,
       long_description_content_type='text/x-rst',
@@ -48,6 +48,6 @@
           'Topic :: Scientific/Engineering :: Bio-Informatics',
           'License :: OSI Approved :: MIT License',
           'Programming Language :: Python :: 2.7',
-          'Programming Language :: Python :: 3.4',
+          'Programming Language :: Python :: 3.5',
       ],
       zip_safe=False)

uncurl/fit_dist_data.py

@@ -12,7 +12,7 @@ def GetDistFitError(Dat):
     #Assumes data to be in the form of a numpy matrix 
     # TODO: make this work for sparse inputs?
     # TODO: fixed number of bins, rather than use the maximum value?
-    # that would 
+    # use np.histogram
     Dat = np.round(Dat).astype(int)
     Dat2 = np.log(1 + Dat)
     BinDat = np.zeros(max(Dat)+1)
@@ -27,6 +27,8 @@ def GetDistFitError(Dat):
 
     #Create a bin of frequencies and generate frequencies based on distr
     for i in range(0,len(BinDat)):
+        # this is EXTREMELY INEFFICIENT!!!!!
+        # n^2 since Dat==i requires iterating through the whole array
         BinDat[i] = sum(Dat==i)
         Poiss[i] = poisson.pmf(i,m)
         Norm[i] = norm.pdf((i-m+1)/std)

uncurl/gap_score.py

@@ -4,7 +4,7 @@
 import numpy as np
 from sklearn.cluster import KMeans
 
-def preproc_data(data, gene_subset=False):
+def preproc_data(data, gene_subset=False, **kwargs):
     """
     basic data preprocessing before running gap score
 
@@ -32,7 +32,7 @@ def calculate_bounding_box(data):
     maxes = data.max(0)
     return mins, maxes
 
-def calculate_gap(data, clustering, km, B=50):
+def calculate_gap(data, clustering, km, B=50, **kwargs):
     """
     See: https://datasciencelab.wordpress.com/2013/12/27/finding-the-k-in-k-means-clustering/