bugfix: leaked semaphore error (#309)

* use config for n_cpu

* rm import

* fix process loop

* unuse mp.spawn

ref. https://discuss.pytorch.org/t/how-to-fix-a-sigsegv-in-pytorch-when-using-distributed-training-e-g-ddp/113518/10

* fix commentout

trainset_preprocess_pipeline_print.py

@@ -115,10 +115,10 @@ class PreProcess:
                     p = multiprocessing.Process(
                         target=self.pipeline_mp, args=(infos[i::n_p],)
                     )
-                    p.start()
                     ps.append(p)
-                for p in ps:
-                    p.join()
+                    p.start()
+                for i in range(n_p):
+                    ps[i].join()
         except:
             println("Fail. %s" % traceback.format_exc())