Added a benchmark for running the chunks code on the list of snps

2 files changed, 10 insertions, 32 deletions

diff --git a/wqflask/wqflask/my_pylmm/pyLMM/chunks.py b/wqflask/wqflask/my_pylmm/pyLMM/chunks.py
index 4c34080c..abeffee7 100644
--- a/wqflask/wqflask/my_pylmm/pyLMM/chunks.py
+++ b/wqflask/wqflask/my_pylmm/pyLMM/chunks.py
@@ -32,7 +32,7 @@ def divide_into_chunks(the_list, number_chunks):
 
     return chunks
 
-def confirm_chunk(original, result):
+def _confirm_chunk(original, result):
     all_chunked = []
     for chunk in result:
         all_chunked.extend(chunk)
@@ -40,7 +40,7 @@ def confirm_chunk(original, result):
     assert original == all_chunked, "You didn't chunk right"
 
 
-def chunk_test(divide_func):
+def _chunk_test(divide_func):
     import random
     random.seed(7)
 
@@ -58,7 +58,7 @@ def chunk_test(divide_func):
             len(the_list), number_chunks, len(result)))
         print("result:", result)
 
-        confirm_chunk(the_list, result)
+        _confirm_chunk(the_list, result)
 
         amount_off = abs(number_chunks - len(result))
         if amount_off == 0:
@@ -75,13 +75,13 @@ def chunk_test(divide_func):
     return number_exact, total_amount_off
 
 
-def main():
+def _main():
     info = dict()
     #funcs = (("sam", sam_divide_into_chunks), ("zach", zach_divide_into_chunks))
     funcs = (("only one", divide_into_chunks),)
     for name, func in funcs:
         start = time.time()
-        number_exact, total_amount_off = chunk_test(func)
+        number_exact, total_amount_off = _chunk_test(func)
         took = time.time() - start
         info[name] = dict(number_exact=number_exact,
                           total_amount_off=total_amount_off,
@@ -90,7 +90,7 @@ def main():
     print("info is:", info)
 
 if __name__ == '__main__':
-    main()
+    _main()
     print("\nConfirming doctests...")
     import doctest
     doctest.testmod()
\ No newline at end of file
diff --git a/wqflask/wqflask/my_pylmm/pyLMM/lmm.py b/wqflask/wqflask/my_pylmm/pyLMM/lmm.py
index 99040d38..24565af8 100644
--- a/wqflask/wqflask/my_pylmm/pyLMM/lmm.py
+++ b/wqflask/wqflask/my_pylmm/pyLMM/lmm.py
@@ -29,6 +29,8 @@ from pprint import pformat as pf
 
 from utility.benchmark import Bench
 
+from wqflask.my_pylmm.pyLMM import chunks
+
 #np.seterr('raise')
 
 def run_human(pheno_vector,
@@ -70,15 +72,14 @@ def run_human(pheno_vector,
         
             
         with Bench("Create list of inputs"):
-           
             inputs = list(plink_input)
             
         with Bench("Divide into chunks"):
-            divide_into_chunks(inputs, 63)
+            results = chunks.divide_into_chunks(inputs, 63)
             
         for snp, this_id in plink_input:
             with Bench("part before association"):
-                if count > 10000:
+                if count > 500:
                     break
                 count += 1
                 
@@ -103,25 +104,6 @@ def run_human(pheno_vector,
     return p_values, t_stats
 
 
-def divide_into_chunks(the_list, number_chunks):
-    """Divides a list into approximately number_chunks smaller lists"""
-    length = len(the_list)
-
-    if length == 0:
-        return [[]]
-    
-    if length <= number_chunks:
-        number_chunks = length
-
-    chunksize = int(math.ceil(length / number_chunks))
-
-    chunks = []
-    for counter in range(0, length, chunksize):
-        chunks.append(the_list[counter:counter+chunksize])
-
-    return chunks
-
-
 def human_association(snp,
                       n,
                       keep,
@@ -624,7 +606,3 @@ class LMM:
        pl.xlabel("Heritability")
        pl.ylabel("Probability of data")
        pl.title(title)
-       
-       
-if __name__ == '__main__':
-    chunk_test()
\ No newline at end of file