Merge remote-tracking branch 'chengmo/fix_ci' into fix_ci

core/model.py

@@ -138,9 +138,7 @@ class ModelBase(object):
             os.environ["FLAGS_communicator_is_sgd_optimizer"] = '0'
 
         if name == "SGD":
-            reg = envs.get_global_env("hyper_parameters.reg", 0.0001)
-            optimizer_i = fluid.optimizer.SGD(
-                lr, regularization=fluid.regularizer.L2DecayRegularizer(reg))
+            optimizer_i = fluid.optimizer.SGD(lr)
         elif name == "ADAM":
             optimizer_i = fluid.optimizer.Adam(lr, lazy_mode=True)
         elif name == "ADAGRAD":

models/rank/dnn/config.yaml

@@ -71,22 +71,6 @@ runner:
   print_interval: 10
   phases: [phase1]
 
-- name: single_gpu_train
-  class: train
-  # num of epochs
-  epochs: 4
-  # device to run training or infer
-  device: gpu
-  selected_gpus: "2"
-  save_checkpoint_interval: 2 # save model interval of epochs
-  save_inference_interval: 4 # save inference
-  save_checkpoint_path: "increment" # save checkpoint path
-  save_inference_path: "inference" # save inference path
-  save_inference_feed_varnames: [] # feed vars of save inference
-  save_inference_fetch_varnames: [] # fetch vars of save inference
-  init_model_path: "" # load model path
-  print_interval: 10
-
 - name: single_cpu_infer
   class: infer
   # num of epochs
@@ -96,33 +80,6 @@ runner:
   init_model_path: "increment/0" # load model path
   phases: [phase2]
 
-- name: local_cluster_cpu_ps_train
-  class: local_cluster
-  epochs: 4
-  device: cpu
-  save_checkpoint_interval: 2 # save model interval of epochs
-  save_inference_interval: 4 # save inference
-  save_checkpoint_path: "increment" # save checkpoint path
-  save_inference_path: "inference" # save inference path
-  save_inference_feed_varnames: [] # feed vars of save inference
-  save_inference_fetch_varnames: [] # fetch vars of save inference
-  init_model_path: "" # load model path
-  print_interval: 1
-
-- name: multi_gpu_train
-  class: train
-  epochs: 4
-  device: gpu
-  selected_gpus: "2,3"
-  save_checkpoint_interval: 2 # save model interval of epochs
-  save_inference_interval: 4 # save inference
-  save_checkpoint_path: "increment" # save checkpoint path
-  save_inference_path: "inference" # save inference path
-  save_inference_feed_varnames: [] # feed vars of save inference
-  save_inference_fetch_varnames: [] # fetch vars of save inference
-  init_model_path: "" # load model path
-  print_interval: 10
-
 # runner will run all the phase in each epoch
 phase:
 - name: phase1