Change folder structure for NAS algorithms, rename NASRL to ENAS (kub…

…eflow#1143) * Init commit * Rename nasrl to enas * Modify docker ignore * Undo changes for nas monitor js * Fix enas cnn docker image name * Rename ENAS in README
sperlingxx · Apr 20, 2020 · 9d2ae1d · 9d2ae1d
1 parent 8cf9cce
commit 9d2ae1d
Show file tree

Hide file tree

Showing 34 changed files with 93 additions and 91 deletions.
diff --git a/.dockerignore b/.dockerignore
@@ -2,7 +2,7 @@
 .gitignore
 docs
 examples
-!examples/v1alpha3/NAS-training-containers/RL-cifar10
+!examples/v1alpha3/nas/enas-cnn-cifar10
 manifests
 pkg/ui/*/frontend/node_modules
 pkg/ui/*/frontend/build
diff --git a/README.md b/README.md
@@ -95,7 +95,7 @@ Currently Katib supports the following exploration algorithms:
 
 #### Neural Architecture Search
 
-* [Reinforcement Learning](https://github.com/kubeflow/katib/tree/master/pkg/suggestion/v1alpha3/NAS_Reinforcement_Learning)
+* [Efficient Neural Architecture Search (ENAS)](https://github.com/kubeflow/katib/tree/master/pkg/suggestion/v1alpha3/nas/enas)
 
 
 ## Components in Katib

diff --git a/cmd/suggestion/nasrl/v1alpha3/Dockerfile → cmd/suggestion/nas/enas/v1alpha3/Dockerfile b/cmd/suggestion/nasrl/v1alpha3/Dockerfile → cmd/suggestion/nas/enas/v1alpha3/Dockerfile
@@ -1,20 +1,20 @@
 FROM python:3.6
 
 RUN if [ "$(uname -m)" = "ppc64le" ]; then \
-        apt-get -y update && \
-        apt-get -y install gfortran libopenblas-dev liblapack-dev && \
-        pip install cython; \
+    apt-get -y update && \
+    apt-get -y install gfortran libopenblas-dev liblapack-dev && \
+    pip install cython; \
     fi
 RUN GRPC_HEALTH_PROBE_VERSION=v0.3.1 && \
     if [ "$(uname -m)" = "ppc64le" ]; then \
-	wget -qO/bin/grpc_health_probe https://github.com/grpc-ecosystem/grpc-health-probe/releases/download/${GRPC_HEALTH_PROBE_VERSION}/grpc_health_probe-linux-ppc64le; \
+    wget -qO/bin/grpc_health_probe https://github.com/grpc-ecosystem/grpc-health-probe/releases/download/${GRPC_HEALTH_PROBE_VERSION}/grpc_health_probe-linux-ppc64le; \
     else \
-	wget -qO/bin/grpc_health_probe https://github.com/grpc-ecosystem/grpc-health-probe/releases/download/${GRPC_HEALTH_PROBE_VERSION}/grpc_health_probe-linux-amd64; \
+    wget -qO/bin/grpc_health_probe https://github.com/grpc-ecosystem/grpc-health-probe/releases/download/${GRPC_HEALTH_PROBE_VERSION}/grpc_health_probe-linux-amd64; \
     fi && \
     chmod +x /bin/grpc_health_probe
 
 ADD . /usr/src/app/github.com/kubeflow/katib
-WORKDIR /usr/src/app/github.com/kubeflow/katib/cmd/suggestion/nasrl/v1alpha3
+WORKDIR /usr/src/app/github.com/kubeflow/katib/cmd/suggestion/nas/enas/v1alpha3
 RUN pip install --no-cache-dir -r requirements.txt
 
 ENV PYTHONPATH /usr/src/app/github.com/kubeflow/katib:/usr/src/app/github.com/kubeflow/katib/pkg/apis/manager/v1alpha3/python:/usr/src/app/github.com/kubeflow/katib/pkg/apis/manager/health/python

diff --git a/...gestion/nasrl/v1alpha3/Dockerfile.aarch64 → ...tion/nas/enas/v1alpha3/Dockerfile.aarch64 b/...gestion/nasrl/v1alpha3/Dockerfile.aarch64 → ...tion/nas/enas/v1alpha3/Dockerfile.aarch64
@@ -44,7 +44,7 @@ RUN pip install 'grpcio==1.23.0' 'protobuf==3.9.1' 'googleapis-common-protos==1.
 COPY --from=build-env /bin/grpc_health_probe /bin/
 
 ADD . /usr/src/app/github.com/kubeflow/katib
-WORKDIR /usr/src/app/github.com/kubeflow/katib/cmd/suggestion/nasrl/v1alpha3
+WORKDIR /usr/src/app/github.com/kubeflow/katib/cmd/suggestion/nas/enas/v1alpha3
 
 ENV PYTHONPATH /usr/src/app/github.com/kubeflow/katib:/usr/src/app/github.com/kubeflow/katib/pkg/apis/manager/v1alpha3/python:/usr/src/app/github.com/kubeflow/katib/pkg/apis/manager/health/python
 

diff --git a/cmd/suggestion/nasrl/v1alpha3/main.py → cmd/suggestion/nas/enas/v1alpha3/main.py b/cmd/suggestion/nasrl/v1alpha3/main.py → cmd/suggestion/nas/enas/v1alpha3/main.py
@@ -4,17 +4,17 @@
 
 from pkg.apis.manager.v1alpha3.python import api_pb2_grpc
 from pkg.apis.manager.health.python import health_pb2_grpc
-from pkg.suggestion.v1alpha3.nasrl_service import NasrlService
+from pkg.suggestion.v1alpha3.nas.enas_service import EnasService
 
 
 _ONE_DAY_IN_SECONDS = 60 * 60 * 24
 DEFAULT_PORT = "0.0.0.0:6789"
 
 
 def serve():
-    print("NAS RL Suggestion Service")
+    print("ENAS Suggestion Service")
     server = grpc.server(futures.ThreadPoolExecutor(max_workers=10))
-    service = NasrlService()
+    service = EnasService()
     api_pb2_grpc.add_SuggestionServicer_to_server(service, server)
     health_pb2_grpc.add_HealthServicer_to_server(service, server)
     server.add_insecure_port(DEFAULT_PORT)

diff --git a/...uggestion/nasrl/v1alpha3/requirements.txt → ...estion/nas/enas/v1alpha3/requirements.txt b/...uggestion/nasrl/v1alpha3/requirements.txt → ...estion/nas/enas/v1alpha3/requirements.txt
diff --git a/docs/new-algorithm-service.md b/docs/new-algorithm-service.md
@@ -8,7 +8,7 @@ The design of Katib follows the `ask-and-tell` pattern:
 
 > They often follow a pattern a bit like this: 1. ask for a new set of parameters 1. walk to the experiment and program in the new parameters 1. observe the outcome of running the experiment 1. walk back to your laptop and tell the optimizer about the outcome 1. go to step 1
 
-When an experiment is created, one algorithm service will be created. Then Katib asks for new sets of parameters via `GetSuggestions` GRPC call. After that, Katib creates new trials according to the sets and observe the outcome. When the trials are finished, Katib tells the metrics of the finished trials to the algorithm, and ask another new sets. 
+When an experiment is created, one algorithm service will be created. Then Katib asks for new sets of parameters via `GetSuggestions` GRPC call. After that, Katib creates new trials according to the sets and observe the outcome. When the trials are finished, Katib tells the metrics of the finished trials to the algorithm, and ask another new sets.
 
 The new algorithm needs to implement `Suggestion` service defined in [api.proto](../pkg/apis/manager/v1alpha3/api.proto). One sample algorithm looks like:
 
@@ -31,8 +31,8 @@ class HyperoptService(
         # Convert the experiment in GRPC request to the search space.
         # search_space example:
         #   HyperParameterSearchSpace(
-        #       goal: MAXIMIZE, 
-        #       params: [HyperParameter(name: param-1, type: INTEGER, min: 1, max: 5, step: 0), 
+        #       goal: MAXIMIZE,
+        #       params: [HyperParameter(name: param-1, type: INTEGER, min: 1, max: 5, step: 0),
         #                HyperParameter(name: param-2, type: CATEGORICAL, list: cat1, cat2, cat3),
         #                HyperParameter(name: param-3, type: DISCRETE, list: 3, 2, 6),
         #                HyperParameter(name: param-4, type: DOUBLE, min: 1, max: 5, step: )]
@@ -41,41 +41,41 @@ class HyperoptService(
         # Convert the trials in GRPC request to the trials in algorithm side.
         # trials example:
         #   [Trial(
-        #       assignment: [Assignment(name=param-1, value=2), 
-        #                    Assignment(name=param-2, value=cat1), 
-        #                    Assignment(name=param-3, value=2), 
+        #       assignment: [Assignment(name=param-1, value=2),
+        #                    Assignment(name=param-2, value=cat1),
+        #                    Assignment(name=param-3, value=2),
         #                    Assignment(name=param-4, value=3.44)],
-        #       target_metric: Metric(name="metric-2" value="5643"), 
-        #       additional_metrics: [Metric(name=metric-1, value=435), 
+        #       target_metric: Metric(name="metric-2" value="5643"),
+        #       additional_metrics: [Metric(name=metric-1, value=435),
         #                            Metric(name=metric-3, value=5643)],
         #   Trial(
         #       assignment: [Assignment(name=param-1, value=3),
         #                    Assignment(name=param-2, value=cat2),
         #                    Assignment(name=param-3, value=6),
         #                    Assignment(name=param-4, value=4.44)],
-        #       target_metric: Metric(name="metric-2" value="3242"), 
-        #       additional_metrics: [Metric(name=metric=1, value=123), 
+        #       target_metric: Metric(name="metric-2" value="3242"),
+        #       additional_metrics: [Metric(name=metric=1, value=123),
         #                            Metric(name=metric-3, value=543)],
         trials = Trial.convert(request.trials)
         #--------------------------------------------------------------
         # Your code here
         # Implment the logic to generate new assignments for the given request number.
         # For example, if request.request_number is 2, you should return:
         # [
-        #   [Assignment(name=param-1, value=3), 
-        #    Assignment(name=param-2, value=cat2), 
-        #    Assignment(name=param-3, value=3), 
+        #   [Assignment(name=param-1, value=3),
+        #    Assignment(name=param-2, value=cat2),
+        #    Assignment(name=param-3, value=3),
         #    Assignment(name=param-4, value=3.22)
         #   ],
-        #   [Assignment(name=param-1, value=4), 
-        #    Assignment(name=param-2, value=cat4), 
-        #    Assignment(name=param-3, value=2), 
+        #   [Assignment(name=param-1, value=4),
+        #    Assignment(name=param-2, value=cat4),
+        #    Assignment(name=param-3, value=2),
         #    Assignment(name=param-4, value=4.32)
         #   ],
         # ]
         list_of_assignments = your_logic(search_space, trials, request.request_number)
         #--------------------------------------------------------------
-        # Convert list_of_assignments to 
+        # Convert list_of_assignments to
         return api_pb2.GetSuggestionsReply(
             trials=Assignment.generate(list_of_assignments)
         )
@@ -223,8 +223,8 @@ Then add a new step in our CI to run the new e2e test case in [test/workflows/co
 ```diff
 // ...
                   {
-                    name: "run-nasrl-e2e-tests",
-                    template: "run-nasrl-e2e-tests",
+                    name: "run-enas-e2e-tests",
+                    template: "run-enas-e2e-tests",
                   },
                   {
                     name: "run-hyperband-e2e-tests",

diff --git a/examples/v1alpha3/README.md b/examples/v1alpha3/README.md
@@ -323,16 +323,16 @@ docker.io/kubeflowkatib/mxnet-mnist
 docker.io/kubeflowkatib/pytorch-mnist
 ```
 
-- Keras cifar10 example for NAS RL with gpu support, [source](https://github.com/kubeflow/katib/blob/master/examples/v1alpha3/NAS-training-containers/RL-cifar10/Dockerfile.cpu).
+- Keras cifar10 CNN example for ENAS with gpu support, [source](https://github.com/kubeflow/katib/blob/master/examples/v1alpha3/nas/enas-cnn-cifar10/Dockerfile.gpu).
 
 ```
-docker.io/kubeflowkatib/nasrl-cifar10-gpu
+docker.io/kubeflowkatib/enas-cnn-cifar10-gpu
 ```
 
-- Keras cifar10 example for NAS RL with cpu support, [source](https://github.com/kubeflow/katib/blob/master/examples/v1alpha3/NAS-training-containers/RL-cifar10/Dockerfile.cpu).
+- Keras cifar10 CNN example for ENAS with cpu support, [source](https://github.com/kubeflow/katib/blob/master/examples/v1alpha3/nas/enas-cnn-cifar10/Dockerfile.cpu).
 
 ```
-docker.io/kubeflowkatib/nasrl-cifar10-cpu
+docker.io/kubeflowkatib/enas-cnn-cifar10-cpu
 ```
 
 - Pytorch operator mnist example, [source](https://github.com/kubeflow/pytorch-operator/blob/master/examples/mnist/mnist.py).

diff --git a/...ning-containers/RL-cifar10/Dockerfile.cpu → ...lpha3/nas/enas-cnn-cifar10/Dockerfile.cpu b/...ning-containers/RL-cifar10/Dockerfile.cpu → ...lpha3/nas/enas-cnn-cifar10/Dockerfile.cpu
@@ -13,10 +13,10 @@ RUN apt-get update && apt-get install -y software-properties-common && \
       wget
 
 ADD . /usr/src/app/github.com/kubeflow/katib
-WORKDIR /usr/src/app/github.com/kubeflow/katib/examples/v1alpha3/NAS-training-containers/RL-cifar10
+WORKDIR /usr/src/app/github.com/kubeflow/katib/examples/v1alpha3/nas/enas-cnn-cifar10
 
 RUN pip3 install --upgrade pip
 RUN pip3 install --upgrade --no-cache-dir -r requirements-cpu.txt
-ENV PYTHONPATH /usr/src/app/github.com/kubeflow/katib/examples/v1alpha3/NAS-training-containers/RL-cifar10
+ENV PYTHONPATH /usr/src/app/github.com/kubeflow/katib/examples/v1alpha3/nas/enas-cnn-cifar10
 
 ENTRYPOINT ["python3", "-u", "RunTrial.py"]
diff --git a/...ning-containers/RL-cifar10/Dockerfile.gpu → ...lpha3/nas/enas-cnn-cifar10/Dockerfile.gpu b/...ning-containers/RL-cifar10/Dockerfile.gpu → ...lpha3/nas/enas-cnn-cifar10/Dockerfile.gpu
@@ -23,10 +23,10 @@ RUN apt-get update && apt-get install -y software-properties-common && \
 
 
 ADD . /usr/src/app/github.com/kubeflow/katib
-WORKDIR /usr/src/app/github.com/kubeflow/katib/examples/v1alpha3/NAS-training-containers/RL-cifar10
+WORKDIR /usr/src/app/github.com/kubeflow/katib/examples/v1alpha3/nas/enas-cnn-cifar10
 
 RUN pip3 install --upgrade pip
 RUN pip3 install --no-cache-dir -r requirements-gpu.txt
-ENV PYTHONPATH /usr/src/app/github.com/kubeflow/katib/examples/v1alpha3/NAS-training-containers/RL-cifar10
+ENV PYTHONPATH /usr/src/app/github.com/kubeflow/katib/examples/v1alpha3/nas/enas-cnn-cifar10
 
 ENTRYPOINT ["python3.5", "-u", "RunTrial.py"]
diff --git a/...containers/RL-cifar10/ModelConstructor.py → .../nas/enas-cnn-cifar10/ModelConstructor.py b/...containers/RL-cifar10/ModelConstructor.py → .../nas/enas-cnn-cifar10/ModelConstructor.py
diff --git a/...raining-containers/RL-cifar10/RunTrial.py → ...v1alpha3/nas/enas-cnn-cifar10/RunTrial.py b/...raining-containers/RL-cifar10/RunTrial.py → ...v1alpha3/nas/enas-cnn-cifar10/RunTrial.py
diff --git a/...ining-containers/RL-cifar10/op_library.py → ...alpha3/nas/enas-cnn-cifar10/op_library.py b/...ining-containers/RL-cifar10/op_library.py → ...alpha3/nas/enas-cnn-cifar10/op_library.py
diff --git a/...ontainers/RL-cifar10/requirements-cpu.txt → ...nas/enas-cnn-cifar10/requirements-cpu.txt b/...ontainers/RL-cifar10/requirements-cpu.txt → ...nas/enas-cnn-cifar10/requirements-cpu.txt
diff --git a/...ontainers/RL-cifar10/requirements-gpu.txt → ...nas/enas-cnn-cifar10/requirements-gpu.txt b/...ontainers/RL-cifar10/requirements-gpu.txt → ...nas/enas-cnn-cifar10/requirements-gpu.txt
diff --git a/examples/v1alpha3/nasjob-example-RL-cpu.yaml → examples/v1alpha3/nas/enas-example-cpu.yaml b/examples/v1alpha3/nasjob-example-RL-cpu.yaml → examples/v1alpha3/nas/enas-example-cpu.yaml
@@ -12,7 +12,7 @@ apiVersion: "kubeflow.org/v1alpha3"
 kind: Experiment
 metadata:
   namespace: kubeflow
-  name: nas-rl-example-cpu
+  name: enas-example-cpu
 spec:
   parallelTrialCount: 2
   maxTrialCount: 3
@@ -22,7 +22,7 @@ spec:
     goal: 0.99
     objectiveMetricName: Validation-Accuracy
   algorithm:
-    algorithmName: nasrl
+    algorithmName: enas
   trialTemplate:
     goTemplate:
       rawTemplate: |-
@@ -36,7 +36,7 @@ spec:
             spec:
               containers:
               - name: {{.Trial}}
-                image: docker.io/kubeflowkatib/nasrl-cifar10-cpu
+                image: docker.io/kubeflowkatib/enas-cnn-cifar10-cpu
                 command:
                 - "python3.5"
                 - "-u"

diff --git a/examples/v1alpha3/nasjob-example-RL-gpu.yaml → examples/v1alpha3/nas/enas-example-gpu.yaml b/examples/v1alpha3/nasjob-example-RL-gpu.yaml → examples/v1alpha3/nas/enas-example-gpu.yaml
@@ -9,7 +9,7 @@ apiVersion: "kubeflow.org/v1alpha3"
 kind: Experiment
 metadata:
   namespace: kubeflow
-  name: nas-rl-example-gpu
+  name: enas-example-gpu
 spec:
   parallelTrialCount: 3
   maxTrialCount: 12
@@ -19,7 +19,7 @@ spec:
     goal: 0.99
     objectiveMetricName: Validation-Accuracy
   algorithm:
-    algorithmName: nasrl
+    algorithmName: enas
   trialTemplate:
     goTemplate:
       rawTemplate: |-
@@ -33,7 +33,7 @@ spec:
             spec:
               containers:
               - name: {{.Trial}}
-                image: docker.io/kubeflowkatib/nasrl-cifar10-gpu
+                image: docker.io/kubeflowkatib/enas-cnn-cifar10-gpu
                 command:
                 - "python3.5"
                 - "-u"

diff --git a/manifests/v1alpha3/katib-controller/katib-config.yaml b/manifests/v1alpha3/katib-controller/katib-config.yaml
@@ -38,8 +38,8 @@ data:
       "tpe": {
         "image": "gcr.io/kubeflow-images-public/katib/v1alpha3/suggestion-hyperopt"
       },
-      "nasrl": {
-        "image": "gcr.io/kubeflow-images-public/katib/v1alpha3/suggestion-nasrl",
+      "enas": {
+        "image": "gcr.io/kubeflow-images-public/katib/v1alpha3/suggestion-enas",
         "imagePullPolicy": "Always",
         "resources": {
           "limits": {

diff --git a/manifests/v1alpha3/katib-controller/trialTemplateConfigmapLabeled.yaml b/manifests/v1alpha3/katib-controller/trialTemplateConfigmapLabeled.yaml
@@ -28,7 +28,7 @@ data:
             {{- end}}
             {{- end}}
           restartPolicy: Never
-  nasRLCPUTemplate: |-
+  enasCPUTemplate: |-
     apiVersion: batch/v1
     kind: Job
     metadata:
@@ -39,7 +39,7 @@ data:
         spec:
           containers:
           - name: {{.Trial}}
-            image: docker.io/kubeflowkatib/nasrl-cifar10-cpu
+            image: docker.io/kubeflowkatib/enas-cnn-cifar10-cpu
             command:
             - "python3.5"
             - "-u"

diff --git a/...inforcement_Learning/AlgorithmSettings.py → ...on/v1alpha3/nas/enas/AlgorithmSettings.py b/...inforcement_Learning/AlgorithmSettings.py → ...on/v1alpha3/nas/enas/AlgorithmSettings.py
diff --git a/.../NAS_Reinforcement_Learning/Controller.py → ...uggestion/v1alpha3/nas/enas/Controller.py b/.../NAS_Reinforcement_Learning/Controller.py → ...uggestion/v1alpha3/nas/enas/Controller.py
diff --git a/...3/NAS_Reinforcement_Learning/Operation.py → ...suggestion/v1alpha3/nas/enas/Operation.py b/...3/NAS_Reinforcement_Learning/Operation.py → ...suggestion/v1alpha3/nas/enas/Operation.py
diff --git a/...pha3/NAS_Reinforcement_Learning/README.md → pkg/suggestion/v1alpha3/nas/enas/README.md b/...pha3/NAS_Reinforcement_Learning/README.md → pkg/suggestion/v1alpha3/nas/enas/README.md
@@ -1,6 +1,8 @@
-# About the Neural Architecture Search with Reinforcement Learning Suggestion
+# About the Efficient Neural Architecture Search
 
-The algorithm follows the idea proposed in _Neural Architecture Search with Reinforcement Learning_ by Zoph & Le (https://arxiv.org/abs/1611.01578), and the implementation is based on the github of _Efﬁcient Neural Architecture Search via Parameter Sharing_ (https://github.com/melodyguan/enas and https://github.com/google-research/google-research/tree/master/enas_lm). It uses a recurrent neural network with LSTM cells as controller to generate neural architecture candidates. And this controller network is updated by policy gradients. However, it currently does not support parameter sharing.
+The algorithm follows the idea proposed in _Efficient Neural Architecture Search via Parameter Sharing_ by Hieu Pham, Melody Y. Guan, Barret Zoph, Quoc V. Le and Jeff Dean (https://arxiv.org/abs/1802.03268) and _Neural Architecture Search with Reinforcement Learning_ by Barret Zoph and Quoc V. Le (https://arxiv.org/abs/1611.01578).
+
+The implementation is based on the github of _Efﬁcient Neural Architecture Search via Parameter Sharing_ (https://github.com/melodyguan/enas and https://github.com/google-research/google-research/tree/master/enas_lm). It uses a recurrent neural network with LSTM cells as controller to generate neural architecture candidates. And this controller network is updated by policy gradients. However, it currently does not support parameter sharing.
 
 ## Definition of a Neural Architecture
 
@@ -130,5 +132,5 @@ This neural architecture can be visualized as
 2. Add support for recurrent neural networks and build a training container for the Penn Treebank task.
 3. Add parameter sharing, if possible.
 4. Change LSTM cell from self defined functions in LSTM.py to `tf.nn.rnn_cell.LSTMCell`
-5. Store the suggestion checkpoint to PVC to protect against unexpected nasrl service pod restarts
+5. Store the suggestion checkpoint to PVC to protect against unexpected enas service pod restarts
 6. Add `RequestCount` into API so that the suggestion can clean the information of completed studies.
diff --git a/...a3/NAS_Reinforcement_Learning/example.png → pkg/suggestion/v1alpha3/nas/enas/example.png b/...a3/NAS_Reinforcement_Learning/example.png → pkg/suggestion/v1alpha3/nas/enas/example.png
diff --git a/pkg/suggestion/v1alpha3/nasrl_service.py → pkg/suggestion/v1alpha3/nas/enas_service.py b/pkg/suggestion/v1alpha3/nasrl_service.py → pkg/suggestion/v1alpha3/nas/enas_service.py
@@ -7,13 +7,13 @@
 
 from pkg.apis.manager.v1alpha3.python import api_pb2
 from pkg.apis.manager.v1alpha3.python import api_pb2_grpc
-from pkg.suggestion.v1alpha3.NAS_Reinforcement_Learning.Controller import Controller
-from pkg.suggestion.v1alpha3.NAS_Reinforcement_Learning.Operation import SearchSpace
-from pkg.suggestion.v1alpha3.NAS_Reinforcement_Learning.AlgorithmSettings import parseAlgorithmSettings
+from pkg.suggestion.v1alpha3.nas.enas.Controller import Controller
+from pkg.suggestion.v1alpha3.nas.enas.Operation import SearchSpace
+from pkg.suggestion.v1alpha3.nas.enas.AlgorithmSettings import parseAlgorithmSettings
 from pkg.suggestion.v1alpha3.base_health_service import HealthServicer
 
 
-class NAS_RL_Experiment:
+class EnasExperiment:
     def __init__(self, request, logger):
         self.logger = logger
         self.experiment_name = request.experiment.name
@@ -126,9 +126,9 @@ def print_algorithm_settings(self):
         self.logger.info("")
 
 
-class NasrlService(api_pb2_grpc.SuggestionServicer, HealthServicer):
+class EnasService(api_pb2_grpc.SuggestionServicer, HealthServicer):
     def __init__(self, logger=None):
-        super(NasrlService, self).__init__()
+        super(EnasService, self).__init__()
         self.is_first_run = True
         self.experiment = None
         if logger == None:
@@ -213,7 +213,7 @@ def SetValidateContextError(self, context, error_message):
 
     def GetSuggestions(self, request, context):
         if self.is_first_run:
-            self.experiment = NAS_RL_Experiment(request, self.logger)
+            self.experiment = EnasExperiment(request, self.logger)
         experiment = self.experiment
         if request.request_number > 0:
             experiment.num_trials = request.request_number

diff --git a/pkg/ui/v1alpha3/frontend/src/reducers/nasCreate.js b/pkg/ui/v1alpha3/frontend/src/reducers/nasCreate.js
@@ -5,7 +5,7 @@ const initialState = {
   commonParametersMetadata: [
     {
       name: 'Name',
-      value: 'nasrl-example',
+      value: 'enas-example',
       description: 'A name of an experiment',
     },
     {
@@ -50,8 +50,8 @@ const initialState = {
     },
   ],
   additionalMetricNames: [],
-  algorithmName: 'nasrl',
-  allAlgorithms: ['nasrl'],
+  algorithmName: 'enas',
+  allAlgorithms: ['enas'],
   algorithmSettings: [
     {
       name: 'controller_hidden_size',

diff --git a/prow_config.yaml b/prow_config.yaml
@@ -29,7 +29,7 @@ workflows:
     - cmd/suggestion/chocolate/v1alpha3/*
     - cmd/suggestion/hyperband/v1alpha3/*
     - cmd/suggestion/hyperopt/v1alpha3/*
-    - cmd/suggestion/nasrl/v1alpha3/*
+    - cmd/suggestion/nas/enas/v1alpha3/*
     - cmd/suggestion/skopt/v1alpha3/*
     - cmd/ui/v1alpha3/*
     - examples/v1alpha3/*.yaml
@@ -69,7 +69,7 @@ workflows:
     - cmd/suggestion/chocolate/v1alpha3/*
     - cmd/suggestion/hyperband/v1alpha3/*
     - cmd/suggestion/hyperopt/v1alpha3/*
-    - cmd/suggestion/nasrl/v1alpha3/*
+    - cmd/suggestion/nas/enas/v1alpha3/*
     - cmd/suggestion/skopt/v1alpha3/*
     - cmd/ui/v1alpha3/*
     - examples/v1alpha3/*.yaml

diff --git a/scripts/v1alpha3/build.sh b/scripts/v1alpha3/build.sh
@@ -72,8 +72,8 @@ docker build -t ${REGISTRY}/${PREFIX}/suggestion-hyperopt:${TAG} -f ${CMD_PREFIX
 docker build -t ${REGISTRY}/${PREFIX}/suggestion-skopt:${TAG} -f ${CMD_PREFIX}/suggestion/skopt/v1alpha3/Dockerfile .
 docker build -t ${REGISTRY}/${PREFIX}/suggestion-chocolate:${TAG} -f ${CMD_PREFIX}/suggestion/chocolate/v1alpha3/Dockerfile .
 if [ $MACHINE_ARCH == "aarch64" ]; then
-	docker build -t ${REGISTRY}/${PREFIX}/suggestion-nasrl:${TAG} -f ${CMD_PREFIX}/suggestion/nasrl/v1alpha3/Dockerfile.aarch64 .
+	docker build -t ${REGISTRY}/${PREFIX}/suggestion-enas:${TAG} -f ${CMD_PREFIX}/suggestion/nas/enas/v1alpha3/Dockerfile.aarch64 .
 else
-	docker build -t ${REGISTRY}/${PREFIX}/suggestion-nasrl:${TAG} -f ${CMD_PREFIX}/suggestion/nasrl/v1alpha3/Dockerfile .
+	docker build -t ${REGISTRY}/${PREFIX}/suggestion-enas:${TAG} -f ${CMD_PREFIX}/suggestion/nas/enas/v1alpha3/Dockerfile .
 fi
 docker build -t ${REGISTRY}/${PREFIX}/suggestion-hyperband:${TAG} -f ${CMD_PREFIX}/suggestion/hyperband/v1alpha3/Dockerfile .