job.load_trainer_job(job_path, trainer_id)
job._scheduler_ep = "127.0.0.1:9091" # Inform the scheduler IP to trainer
trainer = FLTrainerFactory().create_fl_trainer(job)
trainer._current_ep = "127.0.0.1:{}".format(9000 + trainer_id)
place = fluid.CPUPlace()
trainer.start(place)
r = Gru4rec_Reader()
train_reader = r.reader(train_file_dir, place, batch_size=125)
output_folder = "model_node4"
epoch_i = 0
while not trainer.stop():
epoch_i += 1
train_step = 0
for data in train_reader():
#print(np.array(data['src_wordseq']))
ret_avg_cost = trainer.run(feed=data, fetch=["mean_0.tmp_0"])
train_step += 1
if train_step == trainer._step:
break
avg_ppl = np.exp(ret_avg_cost[0])
newest_ppl = np.mean(avg_ppl)
print("{} Epoch {} start train, train_step {}, ppl {}".format (time.strftime('%Y-%m-%d %H:%M:%S',time.localtime(time.time())), epoch_i, train_step, newest_ppl))
save_dir = (output_folder + "/epoch_%d") % epoch_i
if trainer_id == 0:
print("start save")
trainer.save_inference_program(save_dir)
if epoch_i >= 5:
break
print("({0}, {1})-DP".format(E, delta))
output_folder = "model_node%d" % trainer_id
epoch_id = 0
step = 0
while not trainer.stop():
epoch_id += 1
if epoch_id > 10:
break
print("{} Epoch {} start train".format(
time.strftime('%Y-%m-%d %H:%M:%S', time.localtime(time.time())),
epoch_id))
for step_id, data in enumerate(train_reader()):
acc = trainer.run(feeder.feed(data), fetch=["accuracy_0.tmp_0"])
step += 1
# print("acc:%.3f" % (acc[0]))
acc_val = train_test(train_test_program=test_program,
train_test_reader=test_reader,
train_test_feed=feeder)
print("Test with epoch %d, accuracy: %s" % (epoch_id, acc_val))
compute_privacy_budget(sample_ratio=0.001,
epsilon=0.1,
step=step,
delta=0.00001)
save_dir = (output_folder + "/epoch_%d") % epoch_id
trainer.save_inference_program(output_folder)
