Clean up release tests (#11420)

release/rllib_tests/unit_gpu_tests/cluster.yaml

@@ -0,0 +1,39 @@
+cluster_name: ray-rllib-regression-tests
+
+min_workers: 0
+max_workers: 0
+
+# Cloud-provider specific configuration.
+provider:
+    type: aws
+    region: us-west-2
+    availability_zone: us-west-2a
+    cache_stopped_nodes: False
+
+# How Ray will authenticate with newly launched nodes.
+auth:
+    ssh_user: ubuntu
+
+head_node:
+    InstanceType: p2.xlarge  # Cheaper 1GPU K80 instance
+    ImageId: ami-07728e9e2742b0662 # Deep Learning AMI (Ubuntu 16.04)
+
+    # Set primary volume to 25 GiB
+    BlockDeviceMappings:
+        - DeviceName: /dev/sda1
+          Ebs:
+              VolumeSize: 100
+
+
+# List of shell commands to run to set up nodes.
+setup_commands: []
+
+# Command to start ray on the head node. You don't need to change this.
+head_start_ray_commands:
+    - source activate tensorflow_p36 && ray stop
+    - ulimit -n 65536; source activate tensorflow_p36 && OMP_NUM_THREADS=1 ray start --head --redis-port=6379 --object-manager-port=8076 --autoscaling-config=~/ray_bootstrap_config.yaml
+
+# Command to start ray on worker nodes. You don't need to change this.
+worker_start_ray_commands:
+    - source activate tensorflow_p36 && ray stop
+    - ulimit -n 65536; source activate tensorflow_p36 && OMP_NUM_THREADS=1 ray start --address=$RAY_HEAD_IP:6379 --object-manager-port=8076

release/rllib_tests/unit_gpu_tests/requirements.txt

@@ -0,0 +1,6 @@
+ray[rllib]
+ray[debug]
+torch==1.6+cu101
+torchvision==0.7.0+cu101
+boto3==1.4.8
+cython==0.29.0

release/rllib_tests/unit_gpu_tests/run.sh

@@ -0,0 +1,53 @@
+#!/usr/bin/env bash
+
+ray_version="" 
+commit=""
+ray_branch=""
+
+for i in "$@"
+do
+echo "$i"
+case "$i" in
+    --ray-version=*)
+    ray_version="${i#*=}"
+
+    ;;
+    --commit=*)
+    commit="${i#*=}"
+    ;;
+    --ray-branch=*)
+    ray_branch="${i#*=}"
+    ;;
+    --workload=*)
+    ;;
+    --help)
+    usage
+    exit
+    ;;
+    *)
+    echo "unknown arg, $i"
+    exit 1
+    ;;
+esac
+done
+
+if [[ $ray_version == "" || $commit == "" || $ray_branch == "" ]]
+then
+    echo "Provide --ray-version, --commit, and --ray-branch"
+    exit 1
+fi
+
+echo "version: $ray_version"
+echo "commit: $commit"
+echo "branch: $ray_branch"
+echo "workload: ignored"
+
+wheel="https://s3-us-west-2.amazonaws.com/ray-wheels/$ray_branch/$commit/ray-$ray_version-cp36-cp36m-manylinux1_x86_64.whl"
+
+conda uninstall -y terminado
+source activate tensorflow_p36 && pip install -U pip
+source activate tensorflow_p36 && pip install -U "$wheel"
+
+# Run all test cases, but with a forced num_gpus=1.
+# TODO: (sven) chose correct dir and run over all RLlib tests and example scripts!
+source activate tensorflow_p36 && export RAY_FORCE_NUM_GPUS=1 && cd ~ && python -m pytest test_attention_net_learning.py