added a meta script to launch many comparaison of blas with one command.

theano/misc/check_blas.py

@@ -10,20 +10,23 @@ GTX 470 7.22s
 GTX 285, 6.84s
 GTX 480 5.83s
 """
+import sys
 
 import theano,numpy,time
 import theano.tensor as T
+
 shapes=(2000,2000)
 iters = 10
 
 
-def execute(verbose=True):
+def execute(execute=True, verbose=True):
 
     a=theano.shared(numpy.ones(shapes, dtype=theano.config.floatX))
     b=theano.shared(numpy.ones(shapes, dtype=theano.config.floatX))
     c=theano.shared(numpy.ones(shapes, dtype=theano.config.floatX))
 
     f=theano.function([],updates={c:0.4*c+.8*T.dot(a,b)})
+
     if verbose:
         print 'Some theano flags:'
         print '    blas.ldflags=',theano.config.blas.ldflags
@@ -43,12 +46,15 @@ def execute(verbose=True):
         else:
             print 'ERROR, not able to tell if theano used the cpu or the gpu'
             print f.maker.env.toposort()
+    t0=0
+    t1=-1
 
-    t0=time.time()
-    for i in range(iters):
-        f()
-    t1=time.time()
-    if verbose:
+    if execute:
+        t0=time.time()
+        for i in range(iters):
+            f()
+        t1=time.time()
+    if verbose and execute:
         print
         print 'this execution time took %.2fs'%(t1-t0)
     return t1-t0
@@ -63,26 +69,38 @@ def test():
     
 
 if __name__ == "__main__":
-    execute()
-    print """
+    verbose = True
+    print_only = False
+
+    if '--quiet' in sys.argv:
+        verbose = False
+    if '--print_only' in sys.argv:
+        print_only = True
+
+    t = execute(not print_only, verbose)
+
+    if verbose:
+        print """
  	Some result that you can compare again. They where 10 executions of gemm in float64 with matrix of shape 2000x2000 on FC9.
  	
-        We tested 3 cpus: Xeon E5345, Xeon E5430 and Xeon E5450
+        Cpu tested: Xeon E5345, Xeon E5430, Xeon E5450, Core 2 E8500, Core i7 930(hyper-threads enabled)
 
         Lib tested:
             * numpy with ATLAS from distribution(FC9) package (1 thread)
             * manually compiled numpy and ATLAS with 2 threads
             * goto with 1, 2, 4 and 8 threads.
+                          Xeon   Xeno   Xeon  Core2 i7
+ 	lib/nb threads    E5345  E5430  E5450 E8500 930
 
- 	lib/nb threads    E5345(s)  E5430(s)  E5450(s)
-
-        numpy_FC9_atlas/1 39.2s     35.0s     30.7s
- 	goto/1            18.7s     16.1s     14.2s
-        numpy_MAN_atlas/2 12.0s     11.6s     10.2s
- 	goto/2            9.5s      8.1s      7.1s
- 	goto/4            4.9s      4.4s      3.7s
- 	goto/8            2.7s      2.4s      2.0s
+        numpy_FC9_atlas/1 39.2s  35.0s  30.7s 29.6s 21.5s
+ 	goto/1            18.7s  16.1s  14.2s 13.7s 16.1s
+        numpy_MAN_atlas/2 12.0s  11.6s  10.2s 9.2s  9.0s
+ 	goto/2            9.5s   8.1s   7.1s  7.3s  8.1s
+ 	goto/4            4.9s   4.4s   3.7s  -     4.1s
+ 	goto/8            2.7s   2.4s   2.0s  -     4.1s
  	"""
 
-    print 
-    print "We timed",iters,"executions of gemm with matrix of shapes",shapes
+        print 
+        print "We timed",iters,"executions of gemm with matrix of shapes",shapes
+    else:
+        print t

theano/misc/check_blas_many.sh

+#!/bin/bash
+
+python misc/check_blas.py --print_only
+
+cat /proc/cpuinfo |grep "model name" |uniq
+cat /proc/cpuinfo |grep processor
+free
+uname -a
+
+t0=`THEANO_FLAGS=blas.ldflags= OMP_NUM_THREADS=1 time python misc/check_blas.py --quiet`
+t1=`OMP_NUM_THREADS=1 time python misc/check_blas.py --quiet`
+t2=`OMP_NUM_THREADS=2 time python misc/check_blas.py --quiet`
+t4=`OMP_NUM_THREADS=4 time python misc/check_blas.py --quiet`
+t8=`OMP_NUM_THREADS=8 time python misc/check_blas.py --quiet`
+
+echo "numpy gemm took: $t0"
+echo "theano gemm 1 thread took: $t1"
+echo "theano gemm 2 thread took: $t2"
+echo "theano gemm 4 thread took: $t4"
+echo "theano gemm 8 thread took: $t8"
+
+#Fred to test distro numpy at LISA: PYTHONPATH=/u/bastienf/repos:/usr/lib64/python2.5/site-packages THEANO_FLAGS=blas.ldflags= OMP_NUM_THREADS=8 time python misc/check_blas.py
\ No newline at end of file