def _mini_batch(model, mini_batch, batch, is_train, apply_updates =None ):
global insp_
ce = []
for i in xrange(batch.mini/batch.micro):
if not is_train:
output = model(mini_batch, i)
ce.append([output[0], output[1]])
return _avg(ce)
def _mini_batch(model, mini_batch, batch, is_train, apply_updates=None):
global insp_
ce = []
for i in xrange(batch.mini / batch.micro):
if not is_train:
output = model(mini_batch, i)
ce.append([output[0], output[1]])
return _avg(ce)
def _batch(model, batch_size, batch, is_train=True, apply_updates=None):
ce = []
for i in xrange(batch_size / batch.mini):
ce.append(_mini_batch(model, i, batch, is_train, apply_updates))
return _avg(ce)
time_start_iter = time()
loader.next_train_batch(x_, y_, x_skeleton_)
tr.batch_size = y_.get_value(borrow=True).shape[0]
ce.append(_batch(train_model, tr.batch_size, batch, True, apply_updates))
timing_report(i, time()-time_start_iter, tr.batch_size, res_dir)
print "\t| "+ training_report(ce[-1]) + ", finish total of: 0." + str(i*1.0/loader.n_iter_train)
# End of Epoch
####################################################################
####################################################################
print "\n%s\n\t End of epoch %d, \n printing some debug info.\n%s" \
%('-'*30, epoch, '-'*30)
####################################################################
####################################################################
# print insp_
train_ce.append(_avg(ce))
# validate
valid_ce.append(test_lio_skel(use, test_model, batch, drop, tr.rng, epoch, tr.batch_size, x_, y_, loader, x_skeleton_))
# save best params
# if valid_ce[-1][1] < 0.25:
res_dir = save_results(train_ce, valid_ce, res_dir, params=params)
if not tr.moved: res_dir = move_results(res_dir)
if valid_ce[-1][1] < best_valid:
save_params(params, res_dir, "best")
save_params(params, res_dir)
if valid_ce[-1][1] < best_valid:
best_valid = valid_ce[-1][1]
print('tr.batch_size_after=%d' % tr.batch_size)
ce.append(
_batch(train_model, tr.batch_size, batch, True, apply_updates)[0])
print "the %d iteration,time used:%d" % (i, time() - time_start_iter)
#timing_report(i, time()-time_start_iter, tr.batch_size, res_dir)
print "\t| " + training_report(ce[-1]) + ", finish total of: 0." + str(
i * 1.0 / loader.n_iter_train)
# End of Epoch
####################################################################
####################################################################
print "\n%s\n\t End of epoch %d, \n printing some debug info.\n%s" \
%('-'*30, epoch, '-'*30)
####################################################################
####################################################################
# print insp_
train_ce.append(_avg(ce))
# validate
valid_ce.append(
test_lio_skel(use, test_model, batch, drop, tr.rng, epoch,
tr.batch_size, x_, y_, loader, x_skeleton_))
# save best params
# if valid_ce[-1][1] < 0.25:
res_dir = save_results(train_ce, valid_ce, res_dir, params=params)
if not tr.moved: res_dir = move_results(res_dir)
if valid_ce[-1][1] < best_valid:
save_params(params, res_dir, "best")
save_params(params, res_dir)
if valid_ce[-1][1] < best_valid:
#print out_mean_train, out_std_train
ce.append(ce_temp)
out_mean_train.append(out_mean_temp)
out_std_train.append(out_std_temp)
print "Training: No.%d iter of Total %d, %d s"% (i,loader.n_iter_train, time()-time_start_iter) \
+ "\t| negative_log_likelihood "+ training_report(ce[-1])
# End of Epoch
####################################################################
####################################################################
print "\n%s\n\t End of epoch %d, \n printing some debug info.\n%s" \
%('-'*30, epoch, '-'*30)
####################################################################
####################################################################
print ce
train_ce.append(_avg(ce))
out_mean_all.append(_avg(out_mean_train))
out_std_all.append(_avg(out_std_train))
# validate
valid_ce.append(test_lio_skel(use, test_model, batch, drop, tr.rng, epoch, tr.batch_size, x_, y_, loader, x_skeleton_))
# save best params
res_dir = save_results(train_ce, valid_ce, res_dir, params=params, out_mean_train=out_mean_all,out_std_train=out_std_all)
if not tr.moved: res_dir = move_results(res_dir)
if valid_ce[-1][1] < best_valid:
save_params(params, res_dir, "best")
save_params(params, res_dir)
if valid_ce[-1][1] < best_valid:
best_valid = valid_ce[-1][1]
#load data
time_start_iter = time()
loader.next_train_batch(x_, y_, x_skeleton_)
ce_temp, out_mean_temp, out_std_temp = _batch(train_model, tr.batch_size, batch, True, apply_updates)
ce.append(ce_temp)
out_mean_train.append(out_mean_temp)
out_std_train.append(out_std_temp)
print "Training: No.%d iter of Total %d, %d s"% (i,loader.n_iter_train, time()-time_start_iter) \
+ "\t| negative_log_likelihood "+ training_report(ce[-1])
# End of Epoch
####################################################################
print "\n%s\n\t End of epoch %d, \n printing some debug info.\n%s" \
%('-'*30, epoch, '-'*30)
train_ce.append(_avg(ce))
out_mean_all.append(_avg(out_mean_train))
out_std_all.append(_avg(out_std_train))
# validate
valid_ce.append(test_lio_skel(use, test_model, batch, drop, tr.rng, epoch, tr.batch_size, x_, y_, loader, x_skeleton_))
# save best params
res_dir = save_results(train_ce, valid_ce, res_dir, params=net_convnet3d_grbm_early_fusion.params, out_mean_train=out_mean_all,out_std_train=out_std_all)
if not tr.moved: res_dir = move_results(res_dir)
if valid_ce[-1][1] < best_valid:
save_params(net_convnet3d_grbm_early_fusion.params, res_dir, "best")
save_params(net_convnet3d_grbm_early_fusion.params, res_dir)
if valid_ce[-1][1] < best_valid:
best_valid = valid_ce[-1][1]
def _batch(model, batch_size, batch, is_train=True, apply_updates=None):
ce = []
for i in xrange(batch_size/batch.mini): ce.append(_mini_batch(model, i, batch, is_train, apply_updates))
return _avg(ce)
