fix bug

fleetrec/examples/user_define_trainer.py

@@ -29,11 +29,9 @@ class UserDefineTrainer(TranspileTrainer):
         self.regist_context_processor('train_pass', self.train)
 
     def init(self, context):
-        self.model.net()
-        self.model.metrics()
-        self.model.avg_loss()
+        self.model.train_net()
         optimizer = self.model.optimizer()
-        optimizer.minimize(self.model._cost)
+        optimizer.minimize((self.model.get_cost_op()))
 
         self.fetch_vars = []
         self.fetch_alias = []

fleetrec/run.py

@@ -21,10 +21,15 @@ def set_runtime_envs(cluster_envs, engine_yaml):
 
     if cluster_envs is None:
         cluster_envs = {}
-    cluster_envs.update(cluster_envs)
-    cluster_envs.update(_envs)
-    # envs.set_runtime_envions(cluster_envs)
-    print(envs.pretty_print_envs(cluster_envs, ("Runtime Envs", "Value")))
+    envs.set_runtime_envions(cluster_envs)
+    envs.set_runtime_envions(_envs)
+
+    need_print = {}
+    for k, v in os.environ.items():
+        if k.startswith() == "trainer.":
+            need_print[k] = v
+
+    print(envs.pretty_print_envs(need_print, ("Runtime Envs", "Value")))
 
 
 def get_engine(engine):
@@ -45,6 +50,7 @@ def single_engine(args):
     single_envs = {}
     single_envs["trainer.trainer"] = "SingleTrainer"
     single_envs["trainer.threads"] = "2"
+    single_envs["trainer.engine"] = "single"
     set_runtime_envs(single_envs, args.engine_extras)
     trainer = TrainerFactory.create(args.model)
     return trainer
@@ -55,6 +61,7 @@ def cluster_engine(args):
 
     cluster_envs = {}
     cluster_envs["trainer.trainer"] = "ClusterTrainer"
+    cluster_envs["trainer.engine"] = "cluster"
     set_runtime_envs(cluster_envs, args.engine_extras)
 
     envs.set_runtime_envions(cluster_envs)
@@ -85,6 +92,7 @@ def local_cluster_engine(args):
     cluster_envs["trainer.trainer"] = "ClusterTrainer"
     cluster_envs["trainer.strategy"] = "async"
     cluster_envs["trainer.threads"] = "2"
+    cluster_envs["trainer.engine"] = "local_cluster"
     cluster_envs["CPU_NUM"] = "2"
 
     set_runtime_envs(cluster_envs, args.engine_extras)
@@ -102,8 +110,12 @@ def local_mpi_engine(args):
     mpi = util.run_which("mpirun")
     if not mpi:
         raise RuntimeError("can not find mpirun, please check environment")
+    cluster_envs = {}
+    cluster_envs["mpirun"] = mpi
+    cluster_envs["trainer.trainer"] = "CtrCodingTrainer"
+    cluster_envs["log_dir"] = "logs"
+    cluster_envs["trainer.engine"] = "local_cluster"
 
-    cluster_envs = {"mpirun": mpi, "trainer.trainer": "CtrCodingTrainer", "log_dir": "logs"}
     set_runtime_envs(cluster_envs, args.engine_extras)
     launch = LocalMPIEngine(cluster_envs, args.model)
     return launch