diff --git a/benchmark/cluster/vgg16/fluid/README.md b/benchmark/cluster/vgg16/fluid/README.md index 02b17dceb9e235fca321a346cac2cf2df4ddafd5..71a3a934d20b0328ec41dbc34ca3b384749ca49a 100644 --- a/benchmark/cluster/vgg16/fluid/README.md +++ b/benchmark/cluster/vgg16/fluid/README.md @@ -13,4 +13,3 @@ Check the logs for the distributed training progress and analyze the performance ## Enable verbos logs Edit `pserver.yaml` and `trainer.yaml` and add an environment variable `GLOG_v=3` to see what happend in detail. - diff --git a/benchmark/cluster/vgg16/fluid/vgg16.py b/benchmark/cluster/vgg16/fluid/vgg16.py index 3c7b5bf2f14efb00f670f19f3ff6ca2d2116b208..88d6d79cc063d87f4c5b4bf8bb4df67dc3dfe6f8 100644 --- a/benchmark/cluster/vgg16/fluid/vgg16.py +++ b/benchmark/cluster/vgg16/fluid/vgg16.py @@ -11,7 +11,6 @@ # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. - """VGG16 benchmark in Fluid""" from __future__ import print_function diff --git a/benchmark/cluster/vgg16/v2/pserver.yaml b/benchmark/cluster/vgg16/v2/pserver.yaml index ed1671bbbd458a3bdc82bbd44dce108182500c6a..943675e147212ebf9b2007b9f914bdc8d6d2ba4e 100644 --- a/benchmark/cluster/vgg16/v2/pserver.yaml +++ b/benchmark/cluster/vgg16/v2/pserver.yaml @@ -1,13 +1,13 @@ apiVersion: extensions/v1beta1 kind: ReplicaSet metadata: - name: vgg16job-pserver + name: vgg16v2job-pserver spec: replicas: 10 template: metadata: labels: - paddle-job-pserver: vgg16job + paddle-job-pserver: vgg16v2job spec: hostNetwork: true imagePullSecrets: @@ -21,7 +21,7 @@ spec: containerPort: 30236 env: - name: PADDLE_JOB_NAME - value: vgg16job + value: vgg16v2job - name: TRAINERS value: "20" - name: PSERVERS diff --git a/benchmark/cluster/vgg16/v2/trainer.yaml b/benchmark/cluster/vgg16/v2/trainer.yaml index 75fffc64b009839d8e561430589e5bc467cfec36..200b6dc304a5d73c80ca54f5d45e987d2e893d68 100644 --- a/benchmark/cluster/vgg16/v2/trainer.yaml +++ b/benchmark/cluster/vgg16/v2/trainer.yaml @@ -1,14 +1,14 @@ apiVersion: batch/v1 kind: Job metadata: - name: vgg16job-trainer + name: vgg16v2job-trainer spec: parallelism: 20 completions: 20 template: metadata: labels: - paddle-job: vgg16job + paddle-job: vgg16v2job spec: imagePullSecrets: - name: job-registry-secret @@ -20,7 +20,7 @@ spec: command: ["paddle_k8s", "start_trainer", "v2"] env: - name: PADDLE_JOB_NAME - value: vgg16job + value: vgg16v2job - name: TRAINERS value: "20" - name: PSERVERS