def test_DBN(dataset, hyper): datasets = dataset.sharedTrain, dataset.sharedValid, dataset.sharedTest train_set_x, train_set_y = dataset.sharedTrain valid_set_x, valid_set_y = dataset.sharedValid test_set_x, test_set_y = dataset.sharedTest n_train_batches = train_set_x.get_value(borrow=True).shape[0] / hyper.batchSize numpy_rng = numpy.random.RandomState(123) print '... building the model' dbn = DBN(numpy_rng=numpy_rng, n_ins=dataset.n_in, hidden_layers_sizes=hyper.nHidden, n_outs=dataset.n_out) print '... getting the pretraining functions' pretraining_fns = dbn.pretraining_functions(train_set_x=train_set_x, batch_size=hyper.batchSize, k=hyper.k) # Start timeout timer wait_timeout(test_DBN, 30) print '... pre-training the model' start_time = time.time() for i in xrange(dbn.n_layers): for epoch in xrange(hyper.pretrainingEpochs): print "Pretraining epoch ", epoch, ", time ", (time.time() - start_time) / 60.0 c = [] for batch_index in xrange(n_train_batches): c.append(pretraining_fns[i](index=batch_index, lr=hyper.pretrainingLearningRate)) print 'Pre-training layer %i, epoch %d, cost ' % (i, epoch), print numpy.mean(c) end_time = time.time() print('The pretraining code for file ' + os.path.split(__file__)[1] + ' ran for %.2fm' % ((end_time - start_time) / 60.)) print '... getting the finetuning functions' train_fn, validate_model, test_model = dbn.build_finetune_functions( datasets=datasets, batch_size=hyper.batchSize, learning_rate=hyper.learningRate) print '... finetunning the model' # early-stopping parameters patience = 4 * n_train_batches # look as this many examples regardless validation_frequency = min(n_train_batches, patience / 2) best_params = None best_validation_loss = numpy.inf test_score = 0. start_time = time.time() done_looping = False epoch = 0 while (epoch < hyper.numberEpochs) and (not done_looping): epoch = epoch + 1 print "Finetuning epoch ", epoch, ", time ", (time.time() - start_time) / 60.0 for minibatch_index in xrange(n_train_batches): minibatch_avg_cost = train_fn(minibatch_index) iter = (epoch - 1) * n_train_batches + minibatch_index if (iter + 1) % validation_frequency == 0: validation_losses = validate_model() this_validation_loss = numpy.mean(validation_losses) print('epoch %i, minibatch %i/%i, validation error %f %%' % \ (epoch, minibatch_index + 1, n_train_batches, this_validation_loss * 100.)) if this_validation_loss < best_validation_loss: if (this_validation_loss < best_validation_loss * hyper.improvementThreshold): patience = max(patience, iter * hyper.patienceIncrease) best_validation_loss = this_validation_loss best_iter = iter test_losses = test_model() test_score = numpy.mean(test_losses) print((' epoch %i, minibatch %i/%i, test error of ' 'best model %f %%') % (epoch, minibatch_index + 1, n_train_batches, test_score * 100.)) if patience <= iter: done_looping = True break end_time = time.time() print(('Optimization complete with best validation score of %f %%,' 'with test performance %f %%') % (best_validation_loss * 100., test_score * 100.)) print('The fine tuning code for file ' + os.path.split(__file__)[1] + ' ran for %.2fm' % ((end_time - start_time) / 60.))
def run_dbn_model(data_set, finetune_lr=0.1, pretraining_epochs=10, pretrain_lr=0.01, k=1, training_epochs=1000, batch_size=10): # get partitioned data sets train_data, test_data, validation_data = data_set.get_data() # get train x,y, and id train_data_x, train_data_y, train_data_id = train_data train_ten_x, train_ten_y = theano.shared(np.asarray(train_data_x, dtype=theano.config.floatX)), \ theano.shared(np.asarray(train_data_y, dtype='int32')) #theano.shared(np.asarray(pd.get_dummies(train_data_y).as_matrix(), dtype='int32')) # get test x,y, and id test_data_x, test_data_y, test_data_id = test_data test_ten_x, test_ten_y = theano.shared(np.asarray(test_data_x, dtype=theano.config.floatX)), \ theano.shared(np.asarray(test_data_y, dtype='int32')) #theano.shared(np.asarray(pd.get_dummies(test_data_y).as_matrix(), dtype='int32')) # get validation x,y and id validation_data_x, validation_data_y, validation_data_id = validation_data validation_ten_x, validation_ten_y = theano.shared(np.asarray(validation_data_x, dtype=theano.config.floatX)), \ theano.shared(np.asarray(validation_data_y, dtype='int32')) #theano.shared(np.asarray(pd.get_dummies(validation_data_y).as_matrix(), dtype='int32')) ten_data_set = [(train_ten_x,train_ten_y), (validation_ten_x, validation_ten_y), (test_ten_x, test_ten_y)] # compute number of minibatches for training, validation and testing n_train_batches = train_ten_x.get_value(borrow=True).shape[0] / batch_size print ("n_train_batches: " + str(n_train_batches)) cols = train_data_x.shape[1] assert(train_data_x.shape[1] == test_data_x.shape[1] and test_data_x.shape[1] == validation_data_x.shape[1]) print("cols: " + str(cols)) print("train_x" + str(train_ten_x.get_value(borrow=True).shape)) print("train_y" + str(train_ten_y.get_value(borrow=True).shape)) print("valid_x" + str(validation_ten_x.get_value(borrow=True).shape)) print("valid_y" + str(validation_ten_y.get_value(borrow=True).shape)) print("test_x" + str(test_ten_x.get_value(borrow=True).shape)) print("test_y" + str(test_ten_y.get_value(borrow=True).shape)) # numpy random generator numpy_rng = np.random.RandomState(123) print '... building the model' # construct the Deep Belief Network dbn = DBN(numpy_rng=numpy_rng, n_ins= cols, hidden_layers_sizes=[cols*10, cols*10, cols*10], n_outs=5) # start-snippet-2 ######################### # PRETRAINING THE MODEL # ######################### print '... getting the pretraining functions' pretraining_fns = dbn.pretraining_functions(train_set_x=train_ten_x, batch_size=batch_size, k=k) print '... pre-training the model' start_time = timeit.default_timer() ## Pre-train layer-wise for i in range(dbn.n_layers): # go through pretraining epochs for epoch in range(pretraining_epochs): # go through the training set c = [] for batch_index in range(n_train_batches): c.append(pretraining_fns[i](index=batch_index, lr=pretrain_lr)) print 'Pre-training layer %i, epoch %d, cost ' % (i, epoch), print np.mean(c) end_time = timeit.default_timer() # end-snippet-2 print >> sys.stderr, ('The pretraining code for file ' + os.path.split(__file__)[1] + ' ran for %.2fm' % ((end_time - start_time) / 60.)) ######################## # FINETUNING THE MODEL # ######################## # get the training, validation and testing function for the model print '... getting the finetuning functions' train_fn, validate_model, test_model = dbn.build_finetune_functions( datasets=ten_data_set, batch_size=batch_size, learning_rate=finetune_lr ) print '... finetuning the model' # early-stopping parameters patience = 4 * n_train_batches # look as this many examples regardless patience_increase = 2. # wait this much longer when a new best is # found improvement_threshold = 0.995 # a relative improvement of this much is # considered significant validation_frequency = min(n_train_batches, patience / 2) # go through this many # minibatches before checking the network # on the validation set; in this case we # check every epoch best_validation_loss = np.inf test_score = 0. start_time = timeit.default_timer() done_looping = False epoch = 0 while (epoch < training_epochs) and (not done_looping): epoch = epoch + 1 for minibatch_index in range(n_train_batches): minibatch_avg_cost = train_fn(minibatch_index) iter = (epoch - 1) * n_train_batches + minibatch_index if (iter + 1) % validation_frequency == 0: validation_losses = validate_model() this_validation_loss = np.mean(validation_losses) print( 'epoch %i, minibatch %i/%i, validation error %f %%' % ( epoch, minibatch_index + 1, n_train_batches, this_validation_loss * 100. ) ) # if we got the best validation score until now if this_validation_loss < best_validation_loss: #improve patience if loss improvement is good enough if ( this_validation_loss < best_validation_loss * improvement_threshold ): patience = max(patience, iter * patience_increase) # save best validation score and iteration number best_validation_loss = this_validation_loss best_iter = iter # test it on the test set test_losses = test_model() test_score = np.mean(test_losses) print((' epoch %i, minibatch %i/%i, test error of ' 'best model %f %%') % (epoch, minibatch_index + 1, n_train_batches, test_score * 100.)) # if patience <= iter: # done_looping = True # break end_time = timeit.default_timer() print( ( 'Optimization complete with best validation score of %f %%, ' 'obtained at iteration %i, ' 'with test performance %f %%' ) % (best_validation_loss * 100., best_iter + 1, test_score * 100.) ) print >> sys.stderr, ('The fine tuning code for file ' + os.path.split(__file__)[1] + ' ran for %.2fm' % ((end_time - start_time) / 60.))
def test_DBN(finetune_lr=0.05, pretraining_epochs=20, pretrain_lr=0.01, k=1, training_epochs=1000, batch_size=10,DEBUG=0): # datasets = LF.load_cifar() # IMAGE_SIZE = 32 # if DEBUG: # N1 = 400 # N2 = 600 # else: # N1 = 40000 # N2 = 50000 # x, y = datasets[0] # train_set_x = x[:N1] # train_set_y = y[:N1] # valid_set_x = x[N1:N2] # valid_set_y = y[N1:N2] # test_set_x, test_set_y = datasets[1] # datasets = [(train_set_x, train_set_y), # (valid_set_x, valid_set_y), # (test_set_x, test_set_y)] datasets = load_data(DEBUG=DEBUG) IMAGE_SIZE = 28 if DEBUG: N1 = 400 N2 = 600 else: N1 = 40000 N2 = 50000 x, y = datasets[0] train_set_x = x[:N1] train_set_y = y[:N1] valid_set_x = x[N1:N2] valid_set_y = y[N1:N2] test_set_x, test_set_y = datasets[1] datasets = [(train_set_x, train_set_y), (valid_set_x, valid_set_y), (test_set_x, test_set_y)] # compute number of minibatches for training, validation and testing n_train_batches = train_set_x.shape[0].eval()/batch_size n_valid_batches = valid_set_x.shape[0].eval()/batch_size n_test_batches = test_set_x.shape[0].eval()/batch_size print '... building the model' # construct the Deep Belief Network dbn = DBN(n_ins=IMAGE_SIZE**2, hidden_layers_sizes=[500, 500], n_outs=10) ######################### # PRETRAINING THE MODEL # ######################### print '... getting the pretraining functions' pretraining_fns = dbn.pretraining_functions(train_set_x=train_set_x, batch_size=batch_size, k=k) print '... pre-training the model' start_time = timeit.default_timer() ## Pre-train layer-wise for i in xrange(dbn.n_layers): # go through pretraining epochs for epoch in xrange(pretraining_epochs): # go through the training set c = [] for batch_index in xrange(n_train_batches): c.append(pretraining_fns[i](index=batch_index, lr=pretrain_lr)) print 'Pre-training layer %i, epoch %d, cost %.4f' % (i+1, epoch+1, numpy.mean(c)) end_time = timeit.default_timer() print 'The pretraining for DBN ran for %.2fm' % ((end_time-start_time)/60.)) ######################## # FINETUNING THE MODEL # ######################## # get the training, validation and testing function for the model print '... getting the finetuning functions' train_fn, validate_fn, test_fn = dbn.build_finetune_functions( datasets=datasets, batch_size=batch_size, learning_rate=finetune_lr
def train_dbn( datasets, finetune_lr=0.1, pretraining_epochs=100, pretrain_lr=0.01, k=1, training_epochs=1000, batch_size=10 ): train_set_x, train_set_y = datasets[0] valid_set_x, valid_set_y = datasets[1] test_set_x, test_set_y = datasets[2] # compute number of minibatches for training, validation and testing n_train_batches = train_set_x.get_value(borrow=True).shape[0] / batch_size # numpy random generator numpy_rng = np.random.RandomState(123) logging.info("... building the model") # construct the Deep Belief Network dbn = DBN( numpy_rng=numpy_rng, n_ins=train_set_x.get_value(borrow=True).shape[1], hidden_layers_sizes=[200, 200, 200], n_outs=2, ) ######################### # PRETRAINING THE MODEL # ######################### logging.info("... getting the pretraining functions") pretraining_fns = dbn.pretraining_functions(train_set_x=train_set_x, batch_size=batch_size, k=k) logging.info("... pre-training the model") start_time = time.clock() ## Pre-train layer-wise for i in xrange(dbn.n_layers): # go through pretraining epochs for epoch in xrange(pretraining_epochs): # go through the training set c = [] for batch_index in xrange(n_train_batches): c.append(pretraining_fns[i](index=batch_index, lr=pretrain_lr)) logging.info("Pre-training layer %i, epoch %d, cost " % (i, epoch)) logging.info(np.mean(c)) end_time = time.clock() logging.warn( "The pretraining code for file " + os.path.split(__file__)[1] + " ran for %.2fm" % ((end_time - start_time) / 60.0) ) ######################## # FINETUNING THE MODEL # ######################## # get the training, validation and testing function for the model logging.info("... getting the finetuning functions") train_fn, validate_model, test_model, validate_auc, test_auc = dbn.build_finetune_functions( datasets=datasets, batch_size=batch_size, learning_rate=finetune_lr ) logging.info("... finetunning the model") # early-stopping parameters patience = 10 * n_train_batches # look as this many examples regardless patience_increase = 2.0 # wait this much longer when a new best is # found improvement_threshold = 0.995 # a relative improvement of this much is # considered significant validation_frequency = min(n_train_batches, patience / 2) # go through this many # minibatche before checking the network # on the validation set; in this case we # check every epoch best_params = None best_validation_auc = 0 test_score = 0.0 start_time = time.clock() done_looping = False epoch = 0 while (epoch < training_epochs) and (not done_looping): epoch = epoch + 1 for minibatch_index in xrange(n_train_batches): minibatch_avg_cost = train_fn(minibatch_index) iter = (epoch - 1) * n_train_batches + minibatch_index if (iter + 1) % validation_frequency == 0: validation_losses = validate_model() this_validation_loss = np.mean(validation_losses) validation_auc_score = validate_auc() logging.info( "epoch %i, minibatch %i/%i, validation error %f %%, validation auc %f %%" % ( epoch, minibatch_index + 1, n_train_batches, this_validation_loss * 100.0, validation_auc_score * 100.0, ) ) # if we got the best validation score until now if validation_auc_score > best_validation_auc: # improve patience if loss improvement is good enough if validation_auc_score > best_validation_auc / improvement_threshold: patience = max(patience, iter * patience_increase) # save best validation score and iteration number best_validation_auc = validation_auc_score best_iter = iter # test it on the test set test_losses = test_model() test_score = np.mean(test_losses) test_auc_score = test_auc() logging.info( (" epoch %i, minibatch %i/%i, test error of " "best model %f %%, auc of best model %f %%") % (epoch, minibatch_index + 1, n_train_batches, test_score * 100.0, test_auc_score * 100.0) ) if patience <= iter: done_looping = True break end_time = time.clock() logging.info( ("Optimization complete with best validation auc score of %f %%," "with test auc %f %% (zero-one %f %%)") % (best_validation_auc * 100.0, test_auc_score * 100.0, test_score * 100.0) ) logging.warn( "The fine tuning code for file " + os.path.split(__file__)[1] + " ran for %.2fm" % ((end_time - start_time) / 60.0) ) return dbn
def test_DBN(finetune_lr=0.1, pretraining_epochs=100, pretrain_lr=0.01, k=1, training_epochs=1000, dataset='../mnist.pkl.gz', batch_size=10): """ Demonstrates how to train and test a Deep Belief Network. This is demonstrated on MNIST. :type finetune_lr: float :param finetune_lr: learning rate used in the finetune stage :type pretraining_epochs: int :param pretraining_epochs: number of epoch to do pretraining :type pretrain_lr: float :param pretrain_lr: learning rate to be used during pre-training :type k: int :param k: number of Gibbs steps in CD/PCD :type training_epochs: int :param training_epochs: maximal number of iterations ot run the optimizer :type dataset: string :param dataset: path the the pickled dataset :type batch_size: int :param batch_size: the size of a minibatch """ datasets = load_data(dataset) train_set_x, train_set_y = datasets[0] valid_set_x, valid_set_y = datasets[1] test_set_x, test_set_y = datasets[2] # compute number of minibatches for training, validation and testing n_train_batches = train_set_x.get_value(borrow=True).shape[0] // batch_size # numpy random generator numpy_rng = numpy.random.RandomState(123) print('... building the model') # construct the Deep Belief Network dbn = DBN(numpy_rng=numpy_rng, n_ins=28 * 28, hidden_layers_sizes=[1000, 1000, 1000], n_outs=10) # start-snippet-2 ######################### # PRETRAINING THE MODEL # ######################### print('... getting the pretraining functions') pretraining_fns = dbn.pretraining_functions(train_set_x=train_set_x, batch_size=batch_size, k=k) print('... pre-training the model') start_time = timeit.default_timer() # Pre-train layer-wise for i in range(dbn.n_layers): # go through pretraining epochs for epoch in range(pretraining_epochs): # go through the training set c = [] for batch_index in range(n_train_batches): c.append(pretraining_fns[i](index=batch_index, lr=pretrain_lr)) print('Pre-training layer %i, epoch %d, cost ' % (i, epoch), end=' ') print(numpy.mean(c, dtype='float64')) end_time = timeit.default_timer() # end-snippet-2 print('The pretraining code for file ' + os.path.split(__file__)[1] + ' ran for %.2fm' % ((end_time - start_time) / 60.), file=sys.stderr) ######################## # FINETUNING THE MODEL # ######################## # get the training, validation and testing function for the model print('... getting the finetuning functions') train_fn, validate_model, test_model = dbn.build_finetune_functions( datasets=datasets, batch_size=batch_size, learning_rate=finetune_lr) print('... finetuning the model') # early-stopping parameters # look as this many examples regardless patience = 4 * n_train_batches # wait this much longer when a new best is found patience_increase = 2. # a relative improvement of this much is considered significant improvement_threshold = 0.995 # go through this many minibatches before checking the network on # the validation set; in this case we check every epoch validation_frequency = min(n_train_batches, patience / 2) best_validation_loss = numpy.inf test_score = 0. start_time = timeit.default_timer() done_looping = False epoch = 0 while (epoch < training_epochs) and (not done_looping): epoch = epoch + 1 for minibatch_index in range(n_train_batches): train_fn(minibatch_index) iter = (epoch - 1) * n_train_batches + minibatch_index if (iter + 1) % validation_frequency == 0: validation_losses = validate_model() this_validation_loss = numpy.mean(validation_losses, dtype='float64') print('epoch %i, minibatch %i/%i, validation error %f %%' % (epoch, minibatch_index + 1, n_train_batches, this_validation_loss * 100.)) # if we got the best validation score until now if this_validation_loss < best_validation_loss: # improve patience if loss improvement is good enough if (this_validation_loss < best_validation_loss * improvement_threshold): patience = max(patience, iter * patience_increase) # save best validation score and iteration number best_validation_loss = this_validation_loss best_iter = iter # test it on the test set test_losses = test_model() test_score = numpy.mean(test_losses, dtype='float64') print((' epoch %i, minibatch %i/%i, test error of ' 'best model %f %%') % (epoch, minibatch_index + 1, n_train_batches, test_score * 100.)) if patience <= iter: done_looping = True break end_time = timeit.default_timer() print(('Optimization complete with best validation score of %f %%, ' 'obtained at iteration %i, ' 'with test performance %f %%') % (best_validation_loss * 100., best_iter + 1, test_score * 100.)) print('The fine tuning code for file ' + os.path.split(__file__)[1] + ' ran for %.2fm' % ((end_time - start_time) / 60.), file=sys.stderr)
def run_dbn_model(data_set, finetune_lr=0.1, pretraining_epochs=10, pretrain_lr=0.01, k=1, training_epochs=1000, batch_size=10): # get partitioned data sets train_data, test_data, validation_data = data_set.get_data() # get train x,y, and id train_data_x, train_data_y, train_data_id = train_data train_ten_x, train_ten_y = theano.shared(np.asarray(train_data_x, dtype=theano.config.floatX)), \ theano.shared(np.asarray(train_data_y, dtype='int32')) #theano.shared(np.asarray(pd.get_dummies(train_data_y).as_matrix(), dtype='int32')) # get test x,y, and id test_data_x, test_data_y, test_data_id = test_data test_ten_x, test_ten_y = theano.shared(np.asarray(test_data_x, dtype=theano.config.floatX)), \ theano.shared(np.asarray(test_data_y, dtype='int32')) #theano.shared(np.asarray(pd.get_dummies(test_data_y).as_matrix(), dtype='int32')) # get validation x,y and id validation_data_x, validation_data_y, validation_data_id = validation_data validation_ten_x, validation_ten_y = theano.shared(np.asarray(validation_data_x, dtype=theano.config.floatX)), \ theano.shared(np.asarray(validation_data_y, dtype='int32')) #theano.shared(np.asarray(pd.get_dummies(validation_data_y).as_matrix(), dtype='int32')) ten_data_set = [(train_ten_x, train_ten_y), (validation_ten_x, validation_ten_y), (test_ten_x, test_ten_y)] # compute number of minibatches for training, validation and testing n_train_batches = train_ten_x.get_value(borrow=True).shape[0] / batch_size print("n_train_batches: " + str(n_train_batches)) cols = train_data_x.shape[1] assert (train_data_x.shape[1] == test_data_x.shape[1] and test_data_x.shape[1] == validation_data_x.shape[1]) print("cols: " + str(cols)) print("train_x" + str(train_ten_x.get_value(borrow=True).shape)) print("train_y" + str(train_ten_y.get_value(borrow=True).shape)) print("valid_x" + str(validation_ten_x.get_value(borrow=True).shape)) print("valid_y" + str(validation_ten_y.get_value(borrow=True).shape)) print("test_x" + str(test_ten_x.get_value(borrow=True).shape)) print("test_y" + str(test_ten_y.get_value(borrow=True).shape)) # numpy random generator numpy_rng = np.random.RandomState(123) print '... building the model' # construct the Deep Belief Network dbn = DBN(numpy_rng=numpy_rng, n_ins=cols, hidden_layers_sizes=[cols * 10, cols * 10, cols * 10], n_outs=5) # start-snippet-2 ######################### # PRETRAINING THE MODEL # ######################### print '... getting the pretraining functions' pretraining_fns = dbn.pretraining_functions(train_set_x=train_ten_x, batch_size=batch_size, k=k) print '... pre-training the model' start_time = timeit.default_timer() ## Pre-train layer-wise for i in range(dbn.n_layers): # go through pretraining epochs for epoch in range(pretraining_epochs): # go through the training set c = [] for batch_index in range(n_train_batches): c.append(pretraining_fns[i](index=batch_index, lr=pretrain_lr)) print 'Pre-training layer %i, epoch %d, cost ' % (i, epoch), print np.mean(c) end_time = timeit.default_timer() # end-snippet-2 print >> sys.stderr, ('The pretraining code for file ' + os.path.split(__file__)[1] + ' ran for %.2fm' % ((end_time - start_time) / 60.)) ######################## # FINETUNING THE MODEL # ######################## # get the training, validation and testing function for the model print '... getting the finetuning functions' train_fn, validate_model, test_model = dbn.build_finetune_functions( datasets=ten_data_set, batch_size=batch_size, learning_rate=finetune_lr) print '... finetuning the model' # early-stopping parameters patience = 4 * n_train_batches # look as this many examples regardless patience_increase = 2. # wait this much longer when a new best is # found improvement_threshold = 0.995 # a relative improvement of this much is # considered significant validation_frequency = min(n_train_batches, patience / 2) # go through this many # minibatches before checking the network # on the validation set; in this case we # check every epoch best_validation_loss = np.inf test_score = 0. start_time = timeit.default_timer() done_looping = False epoch = 0 while (epoch < training_epochs) and (not done_looping): epoch = epoch + 1 for minibatch_index in range(n_train_batches): minibatch_avg_cost = train_fn(minibatch_index) iter = (epoch - 1) * n_train_batches + minibatch_index if (iter + 1) % validation_frequency == 0: validation_losses = validate_model() this_validation_loss = np.mean(validation_losses) print('epoch %i, minibatch %i/%i, validation error %f %%' % (epoch, minibatch_index + 1, n_train_batches, this_validation_loss * 100.)) # if we got the best validation score until now if this_validation_loss < best_validation_loss: #improve patience if loss improvement is good enough if (this_validation_loss < best_validation_loss * improvement_threshold): patience = max(patience, iter * patience_increase) # save best validation score and iteration number best_validation_loss = this_validation_loss best_iter = iter # test it on the test set test_losses = test_model() test_score = np.mean(test_losses) print((' epoch %i, minibatch %i/%i, test error of ' 'best model %f %%') % (epoch, minibatch_index + 1, n_train_batches, test_score * 100.)) # if patience <= iter: # done_looping = True # break end_time = timeit.default_timer() print(('Optimization complete with best validation score of %f %%, ' 'obtained at iteration %i, ' 'with test performance %f %%') % (best_validation_loss * 100., best_iter + 1, test_score * 100.)) print >> sys.stderr, ('The fine tuning code for file ' + os.path.split(__file__)[1] + ' ran for %.2fm' % ((end_time - start_time) / 60.))