[add]上传训练benchmark by z00560161

2020-10-19 20:22:23 +08:00
parent 22b83024f5
commit 82522e2f61
1225 changed files with 345421 additions and 0 deletions
@@ -0,0 +1,19 @@
+tensorflow_config:
+    # 基本参数
+    data_url: /home/imagenet_TF/
+    # 1p/8p,epoches设为150
+    epoches: 1
+    epochs_between_evals: 1
+    max_train_steps: 1000
+    batch_size: 128
+
+    # 仅多机执行需要配置: ip1:卡数量1,ip2:卡数量2
+    mpirun_ip: 90.90.176.152:8,90.90.176.154:8
+
+    # docker 镜像名称:版本号
+    docker_image: c73:b02
+
+    # 指定 device id, 多个 id 使用空格分隔, 数量需与 rank_size 相同
+    device_group_1p: 0
+    device_group_2p: 0 1
+    device_group_4p: 0 1 2 3