stats_for_k = {}
if not args['dont_run_syntactic']:
# syntactic update step
augmented_costs = []
costs = []
for block_num in xrange(args['syntactic_blocks_to_run']):
training_block = ngram_reader.training_block(data_rng.random_sample())
block_size = training_block.shape[0]
for count in xrange(block_size):
if count % print_freq == 0:
sys.stdout.write('\rk %i b%i: ngram %d of %d' % (model.k, block_num, count, block_size))
sys.stdout.flush()
train_index = sample_cumulative_discrete_distribution(training_block[:,-1], rng=data_rng)
correct_symbols, error_symbols, ngram_frequency = ngram_reader.contrastive_symbols_from_row(training_block[train_index], rng=data_rng)
augmented_cost, cost = model.update_w(*(list(correct_symbols) + list(error_symbols)))
if not np.isfinite(cost):
print 'single nan detected'
save_model('nan_dump.pkl.gz')
import IPython
IPython.embed()
augmented_costs.append(augmented_cost)
costs.append(cost)
if args['syntactic_blocks_to_run'] > 1:
print
print '%i intermediate mean %f' % (block_num, np.mean(costs[-block_size:]))
print
if not np.isfinite(np.mean(costs)):
print 'nan cost mean detected'
save_model('nan_dump.pkl.gz')
print 'burn complete'
while True:
last_time = time.clock()
model.increase_k()
stats_for_k = {}
if not args['dont_run_syntactic']:
# syntactic update step
augmented_costs = []
costs = []
for block_num in xrange(args['syntactic_blocks_to_run']):
training_block = ngram_reader.training_block(data_rng.random_sample())
block_size = training_block.shape[0]
for datum in syntactic_training_data(data_rng, output='k %i b %i' % (model.k, block_num)):
augmented_cost, cost = model.update_w(*datum)
augmented_costs.append(augmented_cost)
costs.append(cost)
if args['syntactic_blocks_to_run'] > 1:
print
print '%i intermediate mean %f' % (block_num, np.mean(costs[-block_size:]))
print
stats_for_k['syntactic_mean'] = np.mean(costs)
stats_for_k['syntactic_std'] = np.std(costs)
print 'training:'
print 'syntactic mean cost \t%f' % stats_for_k['syntactic_mean']
print 'syntactic std cost \t%f' % stats_for_k['syntactic_std']
stats_for_k['syntactic_mean_augmented'] = np.mean(augmented_costs)
stats_for_k['syntactic_std_augmented'] = np.std(augmented_costs)
print 'syntactic mean augmented cost \t%f' % stats_for_k['syntactic_mean_augmented']
last_time = time.clock()
model.increase_k()
stats_for_k = {}
if not args['dont_run_syntactic']:
# syntactic update step
augmented_costs = []
costs = []
for block_num in xrange(args['syntactic_blocks_to_run']):
training_block = ngram_reader.training_block(
data_rng.random_sample())
block_size = training_block.shape[0]
for datum in syntactic_training_data(data_rng,
output='k %i b %i' %
(model.k, block_num)):
augmented_cost, cost = model.update_w(*datum)
augmented_costs.append(augmented_cost)
costs.append(cost)
if args['syntactic_blocks_to_run'] > 1:
print
print '%i intermediate mean %f' % (
block_num, np.mean(costs[-block_size:]))
print
stats_for_k['syntactic_mean'] = np.mean(costs)
stats_for_k['syntactic_std'] = np.std(costs)
print 'training:'
print 'syntactic mean cost \t%f' % stats_for_k['syntactic_mean']
print 'syntactic std cost \t%f' % stats_for_k['syntactic_std']
stats_for_k['syntactic_mean_augmented'] = np.mean(augmented_costs)
stats_for_k['syntactic_std_augmented'] = np.std(augmented_costs)
costs = []
for block_num in xrange(args['syntactic_blocks_to_run']):
training_block = ngram_reader.training_block(
data_rng.random_sample())
block_size = training_block.shape[0]
for count in xrange(block_size):
if count % print_freq == 0:
sys.stdout.write(
'\rk %i b%i: ngram %d of %d' %
(model.k, block_num, count, block_size))
sys.stdout.flush()
train_index = sample_cumulative_discrete_distribution(
training_block[:, -1], rng=data_rng)
correct_symbols, error_symbols, ngram_frequency = ngram_reader.contrastive_symbols_from_row(
training_block[train_index], rng=data_rng)
augmented_cost, cost = model.update_w(
*(list(correct_symbols) + list(error_symbols)))
if not np.isfinite(cost):
print 'single nan detected'
save_model('nan_dump.pkl.gz')
import IPython
IPython.embed()
augmented_costs.append(augmented_cost)
costs.append(cost)
if args['syntactic_blocks_to_run'] > 1:
print
print '%i intermediate mean %f' % (
block_num, np.mean(costs[-block_size:]))
print
if not np.isfinite(np.mean(costs)):
print 'nan cost mean detected'
