def Save_Simulation_As(self, event=None):
self.canvas.Show(False)
dlg = FileDialog(self,
"Save Simulation As...",
default_dir=os.getcwd() + "/sims/",
wildcard='DAT Files|*.dat|All Files|*.*',
save=1)
result = dlg.ShowModal()
if result == 'ok':
filename = dlg.GetPaths()[0]
else:
filename = None
dlg.Destroy()
if filename:
d = zpickle.load(self.tmpfile)
self.params['save_sim_file'] = filename
d['params'] = self.params
zpickle.save(d, filename)
self.modified = False
self.ResetTitle()
self.canvas.Show(True)
return filename
def __load_sim__(self, lfname):
sim = zpickle.load(lfname)
params = sim['params']
params['save_sim_file'] = self.params['save_sim_file']
params['load_sim_file'] = ''
params['continue'] = False
try:
params['initial_weights'] = sim['weights']
params['initial_moments'] = sim['moments']
except KeyError:
self.params = params
return 1
params['load_sim_file'] = self.tmpfile
params['continue'] = True
sim['params'] = params
self.params = params
zpickle.save(sim, self.tmpfile)
return 0
def run_sim_client(params=None, url='http://localhost:4242'):
from xmlrpclib import ServerProxy
import pickle
server = ServerProxy('http://localhost:4242')
if not params:
params = default_params()
params_bytes = pickle.dumps(params, 0)
if params['load_sim_file']:
fid = open(params['load_sim_file'], 'rb')
bytes = fid.read()
fid.close()
lf_bytes = pickle.dumps(bytes, 0)
else:
lf_bytes = ''
sim_bytes = server.run_sim_server(params_bytes, lf_bytes)
sim = pickle.loads(sim_bytes)
print 'Save: %s' % params['save_sim_file']
zpickle.save(sim, params['save_sim_file'])
return sim
def Save_Simulation_As(self,event=None):
self.canvas.Show(False)
dlg = FileDialog(self, "Save Simulation As...",default_dir=os.getcwd()+"/sims/",
wildcard='DAT Files|*.dat|All Files|*.*',save=1)
result = dlg.ShowModal()
if result == 'ok':
filename = dlg.GetPaths()[0]
else:
filename=None
dlg.Destroy()
if filename:
d=zpickle.load(self.tmpfile)
self.params['save_sim_file']=filename
d['params']=self.params
zpickle.save(d,filename)
self.modified=False
self.ResetTitle()
self.canvas.Show(True)
return filename
def __load_sim__(self,lfname):
sim=zpickle.load(lfname)
params=sim['params']
params['save_sim_file']=self.params['save_sim_file']
params['load_sim_file']=''
params['continue']=False
try:
params['initial_weights']=sim['weights']
params['initial_moments']=sim['moments']
except KeyError:
self.params=params
return 1
params['load_sim_file']=self.tmpfile
params['continue']=True
sim['params']=params
self.params=params
zpickle.save(sim,self.tmpfile)
return 0
def save_sim(sim_params, N, C):
sim_params = deepcopy(sim_params)
sim_params['display'] = None
d = {'sim_params': sim_params, 'N': N, 'C': C}
zpickle.save(d, sim_params['save_sim_file'])
def Save_Simulation(self, event=None):
if not self.modified:
return
sfname = self.params['save_sim_file']
def_sfname = default_params()['save_sim_file']
if sfname == def_sfname:
filename = self.Save_Simulation_As()
else:
filename = sfname
d = zpickle.load(self.tmpfile)
d['params'] = self.params
zpickle.save(d, sfname)
self.modified = False
self.ResetTitle()
return filename
def Save_Simulation(self,event=None):
if not self.modified:
return
sfname=self.params['save_sim_file']
def_sfname=default_params()['save_sim_file']
if sfname==def_sfname:
filename=self.Save_Simulation_As()
else:
filename=sfname
d=zpickle.load(self.tmpfile)
d['params']=self.params
zpickle.save(d,sfname)
self.modified=False
self.ResetTitle()
return filename
def run_sim(params=None,parent=None):
if not params:
params=default_params()
elif isinstance(params,basestring):
fname=params
d=zpickle.load(fname)
params=d['params']
if parent:
save_sim_file=params['tmpfile']
params['display']=1
else:
save_sim_file=params['save_sim_file']
if params['load_sim_file']==params['tmpfile']:
params['load_sim_file']=''
if not save_sim_file:
save_sim_file='sims/untitled.dat'
# Deal the random number seed here
if params['random_seed']=='clock':
numpy.random.seed(None)
else:
numpy.random.seed(params['random_seed'])
params['actual_random_seed']=numpy.random.get_state()
params['random_seed2']=int(round(numpy.random.rand()*1e6))
start_time=time.time()
end_time=None
sim_time=None
num_channels=len(params['pattern_input'])
num_moments=get_num_moments(params)
layer=0
test_stimulus=params['test_stimulus'][layer]['type']
total_num_inputs=0
for p in params['pattern_input']:
total_num_inputs=total_num_inputs+p['num_inputs']
num_neurons=prod(params['num_neurons'])
if params['load_sim_file']:
d=zpickle.load(params['load_sim_file'])
load_sim_params=d['params']
weights=d['weights']
moments=d['moments']
initial_weights=d['weights']
initial_moments=d['moments']
t_mat=None
if params['continue']:
t_mat=d['t_mat']
moments_mat=d['moments_mat']
response_mat=d['response_mat']
weights_mat=d['weights_mat']
if not t_mat:
moments_mat=None
response_mat=None
else:
params['initial_weights']=weights
params['initial_moments']=moments
else:
params['continue']=0
full_mask=[]
for p in params['pattern_input']:
full_mask.append(get_circle_mask(p['num_inputs']))
full_mask=numpy.array(full_mask).ravel()
if not params['initial_weights']:
params['initial_weights']=[numpy.asarray(numpy.random.rand(num_neurons,total_num_inputs)*(params['initial_weight_range'][1]-params['initial_weight_range'][0])+params['initial_weight_range'][0],numpy.float64)]
params['initial_weights'][layer]=params['initial_weights'][layer]*(full_mask.repeat(num_neurons))
initial_weights=params['initial_weights']
if not params['initial_moments']:
params['initial_moments']=[numpy.asarray(numpy.random.rand(num_moments,num_neurons)*(params['initial_moment_range'][1]-params['initial_moment_range'][0])+params['initial_moment_range'][0],numpy.float64)]
initial_moments=params['initial_moments']
moments_mat=numpy.dstack( (initial_moments[0],) )
t_mat=None
if not t_mat: # not loaded from file
t_mat=[0]
start_epoch=1
moments_mat=numpy.dstack( (initial_moments[0],) )
if params['keep_every_epoch']:
weights_mat=[initial_weights]
else:
weights_mat=[]
response_mat=[]
response_var=[]
test_stimulus_type=params['test_stimulus'][layer]['type']
if test_stimulus_type==1: # test OR single
response_var=test_OR_single(params,initial_weights)
if not response_var:
params['test_stimulus'][layer]['type']=0
else:
for r in response_var:
response_mat.append([r[0]])
else:
start_epoch=len(t_mat)
test_stimulus_type=params['test_stimulus'][layer]['type']
response_var=[]
if test_stimulus_type==1: # test OR single
response_var=test_OR_single(params,initial_weights)
if not response_var:
params['test_stimulus'][layer]['type']=0
else:
for r,m in zip(response_var,response_mat):
m.append(r[0])
weights=deepcopy(initial_weights)
moments=deepcopy(initial_moments)
response_var_list=[response_var]
extra_mat=[]
sim={'params':params,
'weights':weights,
'moments':moments,
'moments_mat':moments_mat,
'weights_mat':weights_mat,
'response_mat':response_mat,
'response_var_list':response_var_list,
'initial_weights':weights,
'initial_moments':moments,
't_mat':t_mat,
'start_time':start_time,
'end_time':end_time,
'sim_time':sim_time,
'extra_mat':extra_mat}
params_with_images=deepcopy(params)
# make these point to the same thing
params_with_images['saved_input_vectors']=params['saved_input_vectors']
fix_images_directories(params_with_images)
status('Preemtive Save: %s' % save_sim_file,parent)
zpickle.save(sim,save_sim_file)
w=weights[layer]
m=moments[layer]
t0=time.time()
for epoch in range(start_epoch,start_epoch+params['epoch_number']):
t1=time.time()
extra=train(epoch-start_epoch,params_with_images,w,m)
if extra:
extra_mat.append(extra)
# copy over mask stuff
if epoch==start_epoch:
for i,p in enumerate(params['pattern_input']):
params['pattern_input'][i]['mask']=params_with_images['pattern_input'][i]['mask']
dt1=time.time()-t1
dt0=time.time()-t0
frac=(epoch-start_epoch+1.0)/params['epoch_number']
eta=sec2str(round(dt0/frac-dt0))
status("%.4f...ETA %s" % (dt1,eta),parent)
sim['moments_mat']=numpy.dstack( (sim['moments_mat'],m) )
test_stimulus_type=params['test_stimulus'][layer]['type']
if test_stimulus_type==1: # test OR single
response_var=test_OR_single(params,weights)
response_var_list[0]=response_var
if not response_var:
params['test_stimulus'][layer]['type']=0
else:
for r,mat in zip(response_var,response_mat):
mat.append(r[0])
t_mat.append(t_mat[-1]+params['iter_per_epoch'])
if params['keep_every_epoch']:
weights_mat.append(deepcopy(weights))
if params['display'] and epoch%params['epoch_per_display']==0:
if parent:
parent.Plot(sim)
else:
Plot(sim)
if parent:
parent.Yield()
if parent.Stopping():
break
if dt0>(60*60*20): # every 20 minutes
status('Incremental Save: %s' % (save_sim_file),parent)
zpickle.save(sim,save_sim_file)
end_time=time.time()
sim_time=end_time-start_time
if params['display']:
if parent:
parent.Plot(sim)
else:
Plot(sim)
status('Save: %s' % (save_sim_file),parent)
zpickle.save(sim,save_sim_file)
def save(var, fname):
print "Writing ", fname
zpickle.save(var, fname)
fname = hdf5_fname(fname)
print "Writing ", fname
hdf5_save_images(var, fname)
def run_sim(params=None, parent=None):
if not params:
params = default_params()
elif isinstance(params, basestring):
fname = params
d = zpickle.load(fname)
params = d['params']
if parent:
save_sim_file = params['tmpfile']
params['display'] = 1
else:
save_sim_file = params['save_sim_file']
if params['load_sim_file'] == params['tmpfile']:
params['load_sim_file'] = ''
if not save_sim_file is None and not save_sim_file:
save_sim_file = 'sims/untitled.dat'
# Deal the random number seed here
if params['random_seed'] == 'clock':
numpy.random.seed(None)
else:
numpy.random.seed(params['random_seed'])
params['actual_random_seed'] = numpy.random.get_state()
params['random_seed2'] = int(round(numpy.random.rand() * 1e6))
start_time = time.time()
end_time = None
sim_time = None
num_channels = len(params['pattern_input'])
num_moments = get_num_moments(params)
layer = 0
test_stimulus = params['test_stimulus'][layer]['type']
total_num_inputs = 0
for p in params['pattern_input']:
total_num_inputs = total_num_inputs + p['num_inputs']
num_neurons = prod(params['num_neurons'])
if params['load_sim_file']:
d = zpickle.load(params['load_sim_file'])
load_sim_params = d['params']
weights = d['weights']
moments = d['moments']
individual_moments = d['individual_moments']
initial_weights = d['weights']
initial_moments = d['moments']
initial_individual_moments = d['individual_moments']
t_mat = None
if params['continue']:
t_mat = d['t_mat']
moments_mat = d['moments_mat']
individual_moments_mat = d['individual_moments_mat']
response_mat = d['response_mat']
weights_mat = d['weights_mat']
if not t_mat:
moments_mat = None
individual_moments_mat = None
response_mat = None
else:
params['initial_weights'] = deepcopy(weights)
params['initial_moments'] = deepcopy(moments)
params['initial_individual_moments'] = deepcopy(individual_moments)
initial_weights = params['initial_weights']
initial_moments = params['initial_moments']
initial_individual_moments = params['initial_individual_moments']
else:
params['continue'] = 0
full_mask = get_full_mask(params)
if not params['initial_weights']:
params['initial_weights'] = [
numpy.asarray(
numpy.random.rand(num_neurons, total_num_inputs) *
(params['initial_weight_range'][1] -
params['initial_weight_range'][0]) +
params['initial_weight_range'][0], numpy.float64)
]
# I love broadcasting! full_mask has a length of the number of inputs
params['initial_weights'][
layer] = params['initial_weights'][layer] * full_mask
initial_weights = params['initial_weights']
if not params['initial_moments']:
params['initial_moments'] = [
numpy.asarray(
numpy.random.rand(num_moments, num_neurons) *
(params['initial_moment_range'][1] -
params['initial_moment_range'][0]) +
params['initial_moment_range'][0], numpy.float64)
]
# same size as the weights - make into a list if you need to
# have more of these
params['initial_individual_moments'] = [
numpy.asarray(
numpy.random.rand(num_neurons, total_num_inputs) *
(params['initial_moment_range'][1] -
params['initial_moment_range'][0]) +
params['initial_moment_range'][0], numpy.float64)
]
initial_moments = params['initial_moments']
initial_individual_moments = params['initial_individual_moments']
moments_mat = numpy.dstack((initial_moments[0], ))
t_mat = None
if not t_mat: # not loaded from file
t_mat = [0]
start_epoch = 1
moments_mat = numpy.dstack((initial_moments[0], ))
if params['keep_every_epoch']:
weights_mat = [initial_weights]
individual_moments_mat = [initial_individual_moments]
else:
weights_mat = []
individual_moments_mat = []
response_mat = []
response_var = []
test_stimulus_type = params['test_stimulus'][layer]['type']
if test_stimulus_type == 1: # test OR single
response_var = test_OR_single(params, initial_weights)
if not response_var:
params['test_stimulus'][layer]['type'] = 0
else:
for r in response_var:
response_mat.append([r[0]])
else:
start_epoch = len(t_mat)
test_stimulus_type = params['test_stimulus'][layer]['type']
response_var = []
if test_stimulus_type == 1: # test OR single
response_var = test_OR_single(params, initial_weights)
if not response_var:
params['test_stimulus'][layer]['type'] = 0
else:
for r, m in zip(response_var, response_mat):
m.append(r[0])
weights = deepcopy(initial_weights)
moments = deepcopy(initial_moments)
individual_moments = deepcopy(initial_individual_moments)
response_var_list = [response_var]
extra_mat = []
sim = {
'params': params,
'weights': weights,
'moments': moments,
'individual_moments': individual_moments,
'moments_mat': moments_mat,
'individual_moments_mat': individual_moments_mat,
'weights_mat': weights_mat,
'response_mat': response_mat,
'response_var_list': response_var_list,
'initial_weights': initial_weights,
'initial_moments': initial_moments,
't_mat': t_mat,
'start_time': start_time,
'end_time': end_time,
'sim_time': sim_time,
'extra_mat': extra_mat
}
params_with_images = deepcopy(params)
# make these point to the same thing
params_with_images['saved_input_vectors'] = params['saved_input_vectors']
fix_images_directories(params_with_images, parent)
if not save_sim_file is None:
status('Preemtive Save: %s' % save_sim_file, parent)
zpickle.save(sim, save_sim_file)
w = weights[layer]
m = moments[layer]
ind_m = individual_moments[layer]
t0 = time.time()
last_display_time = t0
extra_input = []
try:
for epoch in range(start_epoch, start_epoch + params['epoch_number']):
t1 = time.time()
extra = train(epoch - start_epoch, params_with_images, w, m, ind_m,
extra_input)
if extra:
extra_mat.append(extra)
# copy over mask stuff
if epoch == start_epoch:
for i, p in enumerate(params['pattern_input']):
params['pattern_input'][i]['mask'] = params_with_images[
'pattern_input'][i]['mask']
dt1 = time.time() - t1
dt0 = time.time() - t0
if (time.time() -
last_display_time) > params['minimum_print_time']:
last_display_time = time.time()
frac = (epoch - start_epoch + 1.0) / params['epoch_number']
eta = sec2str(dt0 / frac - dt0)
status("%.4f...ETA %s" % (dt1, eta), parent)
sim['moments_mat'] = numpy.dstack((sim['moments_mat'], m))
test_stimulus_type = params['test_stimulus'][layer]['type']
if test_stimulus_type == 1: # test OR single
response_var = test_OR_single(params, weights)
response_var_list[0] = response_var
if not response_var:
params['test_stimulus'][layer]['type'] = 0
else:
for r, mat in zip(response_var, response_mat):
mat.append(r[0])
t_mat.append(t_mat[-1] + params['iter_per_epoch'])
if params['keep_every_epoch']:
weights_mat.append(deepcopy(weights))
individual_moments_mat.append(deepcopy(individual_moments))
if params['display'] and epoch % params['epoch_per_display'] == 0:
if parent:
parent.Plot(sim)
else:
Plot(sim)
if parent:
parent.Yield()
if parent.Stopping():
break
if dt0 > (60 * 60 * 20): # every 20 minutes
if not save_sim_file is None:
status('Incremental Save: %s' % (save_sim_file), parent)
zpickle.save(sim, save_sim_file)
except KeyboardInterrupt:
status("Stopping!", parent)
pass
end_time = time.time()
sim_time = end_time - start_time
if params['display']:
if parent:
parent.Plot(sim)
else:
Plot(sim)
tt = sec2str(time.time() - t0)
if save_sim_file is None:
status('Save: None. Total time %s' % (tt), parent)
else:
status('Save: %s. Total time %s' % (save_sim_file, tt), parent)
zpickle.save(sim, save_sim_file)
return sim
def save_params(params,fname):
zpickle.save(params,fname)
def save_params(params, fname):
zpickle.save(params, fname)
for i in range(len(var['im'])):
var['im'][i]=(var['im'][i]-mn)/d*maxval
var['im'][i]=var['im'][i].astype(resolution)
var['im_scale_shift']=[d/maxval, mn]
if __name__=="__main__":
# cat stuff
imfname='catim081604.pics'
if not os.path.exists(imfname):
var_raw=make_raw('cat',show=False)
zpickle.save(var_raw,imfname)
var_raw=zpickle.load('catim081604.pics')
# var_raw['im']=[var_raw['im'][0]] # for debugging: use 1 image
imfname='catim081604_dog13.pics'
if not os.path.exists(imfname):
var=make_dog(var_raw)
set_resolution(var,'uint8')
zpickle.save(var,imfname)
imfname='catim081604_white.pics'
if not os.path.exists(imfname):
var=make_white(var_raw)
set_resolution(var,'uint8')
zpickle.save(var,imfname)
