print in_to_hidden.params print hidden_to_out.params print n.params print n.activate([1, 2]) # Naming your NN print LinearLayer(2).name LinearLayer(2, name='foo') print LinearLayer(2).name # Using Recurrent NN n = RecurrentNetwork() n.addInputModule(LinearLayer(2, name='in')) n.addModule(SigmoidLayer(3, name='hidden')) n.addOutputModule(LinearLayer(1, name='out')) n.addConnection(FullConnection(n['in'], n['hidden'], name='c1')) n.addConnection(FullConnection(n['hidden'], n['out'], name='c2')) # Looks back in time one timestep n.addRecurrentConnection(FullConnection(n['hidden'], n['hidden'], name='c3')) # Using RNN, every steps gets different value of Neron n.sortModules() print n.activate([2, 2]) print n.activate([2, 2]) print n.activate([2, 2]) n.reset() # Clear n and Reset it print n.activate([2, 2]) print n.activate([2, 2])
print n
#Recureent Connection Class -which looks back in time one timestep.
n = RecurrentNetwork()
n.addInputModule(LinearLayer(2, name='in'))
n.addModule(SigmoidLayer(3, name='hidden'))
n.addOutputModule(LinearLayer(1, name='out'))
n.addConnection(FullConnection(n['in'], n['hidden'], name='c1'))
n.addConnection(FullConnection(n['hidden'], n['out'], name='c2'))
n.sortModules()
print n.activate((2, 2))
print n.activate((2, 2))
print n.activate((2, 2))
n.reset()
print n.activate((2, 2))
#######################################
######### Classification with feed forward networks
# ###################### graphical output ########################
#To have a nice dataset for visualization,
# we produce a set of points in 2D belonging to three different classes.
means = [(-1, 0), (2, 4), (3, 1)]
cov = [diag([1, 1]), diag([0.5, 1.2]), diag([1.5, 0.7])]
alldata = ClassificationDataSet(2, 1, nb_classes=3)
for n in xrange(400):
for klass in range(3):
input = multivariate_normal(means[klass], cov[klass])
n.addModule(hiddenLayer) n.addOutputModule(outLayer) in_to_hidden = FullConnection(inLayer, hiddenLayer) hidden_to_out = FullConnection(hiddenLayer, outLayer) n.addConnection(in_to_hidden) n.addConnection(hidden_to_out) # this is required to make the MLP usable n.sortModules() print n.activate((2,2)) # forward pass print 'n.params\n', n.params # all weights # same but for recurrent network n = RecurrentNetwork() n.addInputModule(LinearLayer(2, name='in')) n.addModule(SigmoidLayer(3, name='hidden')) n.addOutputModule(LinearLayer(1, name='out')) n.addConnection(FullConnection(n['in'], n['hidden'], name='c1')) n.addConnection(FullConnection(n['hidden'], n['out'], name='c2')) n.addRecurrentConnection(FullConnection(n['hidden'], n['hidden'], name='c3')) n.sortModules() print n.activate((2,2)) # forward pass print n.activate((2,2)) # forward pass print n.activate((2,2)) # forward pass print n.reset(), '\nafter reset' print n.activate((2,2)) # forward pass
n.addInputModule(inLayer)
n.addModule(hiddenLayer)
n.addModule(biasinUnit)
n.addModule(biasoutUnit)
n.addOutputModule(outLayer)
in_to_hidden = FullConnection(inLayer,hiddenLayer)
bias_to_hidden = FullConnection(biasinUnit,hiddenLayer)
bias_to_out = FullConnection(biasoutUnit,outLayer)
hidden_to_out = FullConnection(hiddenLayer,outLayer)
n.addConnection(in_to_hidden)
n.addConnection(bias_to_hidden)
n.addConnection(bias_to_out)
n.addConnection(hidden_to_out)
n.sortModules()
n.reset()
#read the initail weight values from myparam2.txt
filetoopen = os.path.join(os.getcwd(),'myparam2.txt')
if os.path.isfile(filetoopen):
myfile = open('myparam2.txt','r')
c=[]
for line in myfile:
c.append(float(line))
n._setParameters(c)
else:
myfile = open('myparam2.txt','w')
for i in n.params:
myfile.write(str(i)+'\n')
myfile.close()
def ANN_edge_analysis(a_network, a_gene, a_dataset, boot_val): "Creates and trains a network that is created to reflect the structure of the hypothesized network" regulatory_network = FeedForwardNetwork() # retrievingneeded parameters from the input network data_node_list = get_sub_list_from_network(a_network, a_gene, "gene,TF", 1) # Need to add +1 node to the input layer that represents the "other" control variables # describing network modules to be used inLayer = LinearLayer(len(data_node_list)-1) #hiddenLayer = LinearLayer(len(data_node_list)-1)) outLayer = LinearLayer(1) # Adding layers to network regulatory_network.addInputModule(inLayer) #regulatory_network.addModule(hiddenLayer) regulatory_network.addOutputModule(outLayer) # Adding connections between layers #in_to_hidden = LinearConnection(inLayer,hiddenLayer) #hidden_to_out = FullConnection(hiddenLayer, outLayer) in_to_out = FullConnection(inLayer, outLayer) #regulatory_network.addConnection(in_to_hidden) #regulatory_network.addConnection(hidden_to_out) regulatory_network.addConnection(in_to_out) get_nn_details(regulatory_network) # Other stuff added regulatory_network.sortModules() # Formatting the dataset input_dimention = len(data_node_list)-1 print "in_dimention = ", input_dimention DS = SupervisedDataSet( input_dimention, 1 ) # Adding data, there may be a problem with order here where tfs are not always the same... seems ok though for experiment in a_dataset: tf_list = [] gene_list = [] tf_labels = [] for TF in data_node_list: if TF != a_gene: #print TF, "<---" tf_list.append(experiment[TF]) tf_labels.append(TF) else: #print TF, "<---gene" gene_list.append(experiment[TF]) print tf_list print gene_list if (check_missing_experiments(tf_list) == True) and (check_missing_experiments(gene_list) == True): float_tf_list = [float(i) for i in tf_list] float_gene_list = [float(i) for i in gene_list] DS.appendLinked( float_tf_list, float_gene_list ) print "......" print DS # Training trainer = BackpropTrainer(regulatory_network, momentum=0.1, verbose=True, weightdecay=0.01) trainer.setData(DS) result_list = [] boot_count = 0 while boot_count < boot_val: #trainer.trainEpochs(1000) trainer.trainUntilConvergence(validationProportion=0.25) print regulatory_network this = get_nn_details(regulatory_network) result_list.append(this) regulatory_network.reset() boot_count += 1 print tf_labels print regulatory_network.params print in_to_out.params print inLayer pesos_conexiones(regulatory_network) NetworkWriter.writeToFile(regulatory_network, 'trained_net.xml') return result_list
def ANN_blind_analysis(a_network, a_gene, a_dataset, boot_val):
"Creates and trains a network that is created to reflect the structure of the hypothesized network"
regulatory_network = FeedForwardNetwork()
# retrieving needed parameters from the input network
upper_case_data_node_list = get_sub_list_from_network(a_network, a_gene, "gene,TF", 1)
# to lower case for everything
data_node_list = [x.lower() for x in upper_case_data_node_list]
a_gene = a_gene.lower()
# If the target gene is also a TF, remove it from the list as it will be added
if a_gene in data_node_list: data_node_list.remove(a_gene)
print 'what is in data_node_list:'
print data_node_list
if len(data_node_list) == 0:
print "No connections to " + a_gene + " found."
return [a_gene, '0', '0']
# Check for missing entries in the dataset (DS)
# For the main gene
#print a_gene
#print a_dataset[0].keys()
# Check for missing entries in the dataset (DS)
# For the main gene
if a_gene not in a_dataset[0].keys():
#print 'herp'
return [a_gene, '0', '0']
# For the linked genes
for each_gene in data_node_list:
if each_gene not in a_dataset[0].keys():
data_node_list.remove(each_gene)
if len(data_node_list) == 0:
print "No connections to " + a_gene + " found."
return [a_gene, '0', '0']
print len(data_node_list)
print data_node_list
# Need to add +1 node to the input layer that represents the "other" control variables
# describing network modules to be used
inLayer = LinearLayer(len(data_node_list), name="Input_layer")
hiddenLayer = SigmoidLayer(len(data_node_list) + 1, name="Hidden_sigmoid_layer_1")
outLayer = LinearLayer(1, name="Output_layer")
# Adding layers to network
regulatory_network.addInputModule(inLayer)
regulatory_network.addModule(hiddenLayer)
regulatory_network.addOutputModule(outLayer)
# Adding connections between layers
in_to_hidden = FullConnection(inLayer, hiddenLayer)
hidden_to_out = FullConnection(hiddenLayer, outLayer)
regulatory_network.addConnection(in_to_hidden)
regulatory_network.addConnection(hidden_to_out)
get_nn_details(regulatory_network)
# Other stuff added
regulatory_network.sortModules()
# Formatting the dataset
input_dimention = len(data_node_list)
print "in_dimention = ", input_dimention
DS = SupervisedDataSet( input_dimention, 1 )
# Adding data, there may be a problem with order here where tfs are not always the same... seems ok though
# This may not be the best way, but is needed due to the next for statement
data_node_list.append(a_gene)
print 'node list contains: '
print data_node_list
# This is where the ordered dict needs to be used to link the input name to the input node.
for experiment in a_dataset:
tf_list = []
gene_list = []
tf_labels = []
first_round = True
for TF in data_node_list:
if TF != a_gene:
#print TF, "<---"
tf_list.append(experiment[TF])
if first_round == True:
tf_labels.append(TF)
else:
#print TF, "<---gene"
gene_list.append(experiment[TF])
first_round = False
# View the input data sets
print tf_labels
print tf_list
print gene_list
if (check_missing_experiments(tf_list) == True) and (check_missing_experiments(gene_list) == True):
float_tf_list = [float(i) for i in tf_list]
float_gene_list = [float(i) for i in gene_list]
DS.appendLinked( float_tf_list, float_gene_list )
print "......"
print 'Network before training'
print regulatory_network
pesos_conexiones(regulatory_network)
print regulatory_network.outputerror
#print DS
# Training
trainer = RPropMinusTrainer_Evolved(regulatory_network, verbose=False)
trainer.setData(DS)
result_list = []
best_run_error = 1000
boot_count = 0
while boot_count < boot_val:
print '\n'
print 'Bootstrap round ' + str(boot_count + 1)
trainer.trainEpochs(500)
this = get_nn_details(regulatory_network)
# Corrected error
print trainer.total_error
current_run_error = trainer.total_error
print 'Bootstrap round ' + str(boot_count + 1) + ' error: ' + str(current_run_error)
if abs(current_run_error) < abs(best_run_error):
best_run_error = current_run_error
trained_net_filename = a_gene + '_trained_net.xml'
NetworkWriter.writeToFile(regulatory_network, trained_net_filename)
export_to_gml(regulatory_network, tf_labels, a_gene)
#result_list.append(this)
regulatory_network.reset()
regulatory_network.randomize()
trainer = RPropMinusTrainer_Evolved(regulatory_network, verbose=False)
trainer.setData(DS)
boot_count += 1
#print "TF Labels"
#print tf_labels
#print regulatory_network.params
#print inLayer
#print "Pesos Conexiones"
#pesos_conexiones(regulatory_network)
#print dir(regulatory_network)
#print dir(trainer)
#print 'look here'
#print regulatory_network.outputerror
#print '<><><><><>'
#print dir(regulatory_network['SigmoidLayer-7'])
#print '\n'
#print vars(regulatory_network['SigmoidLayer-7'])
#print '\n'
#print regulatory_network['SigmoidLayer-7'].forward
#print regulatory_network['SigmoidLayer-7'].bufferlist
result_list.append(a_gene)
result_list.append(best_run_error)
result_list.append(len(tf_list))
return result_list
