Run Serve Tests on Windows (#10682)

ci/travis/ci.sh

@@ -137,6 +137,7 @@ test_python() {
   if [ "${OSTYPE}" = msys ]; then
     pathsep=";"
     args+=(
+      python/ray/serve/...
       python/ray/tests/...
       -python/ray/tests:test_advanced_2
       -python/ray/tests:test_advanced_3  # test_invalid_unicode_in_worker_log() fails on Windows

python/ray/serve/examples/doc/tutorial_sklearn.py

@@ -6,6 +6,8 @@ import pickle
 import json
 import numpy as np
 import requests
+import os
+import tempfile
 
 from sklearn.datasets import load_iris
 from sklearn.ensemble import GradientBoostingClassifier
@@ -32,9 +34,13 @@ model.fit(train_x, train_y)
 print("MSE:", mean_squared_error(model.predict(val_x), val_y))
 
 # Save the model and label to file
-with open("/tmp/iris_model_logistic_regression.pkl", "wb") as f:
+MODEL_PATH = os.path.join(tempfile.gettempdir(),
+                          "iris_model_logistic_regression.pkl")
+LABEL_PATH = os.path.join(tempfile.gettempdir(), "iris_labels.json")
+
+with open(MODEL_PATH, "wb") as f:
     pickle.dump(model, f)
-with open("/tmp/iris_labels.json", "w") as f:
+with open(LABEL_PATH, "w") as f:
     json.dump(target_names.tolist(), f)
 # __doc_train_model_end__
 
@@ -42,9 +48,9 @@ with open("/tmp/iris_labels.json", "w") as f:
 # __doc_define_servable_begin__
 class BoostingModel:
     def __init__(self):
-        with open("/tmp/iris_model_logistic_regression.pkl", "rb") as f:
+        with open(MODEL_PATH, "rb") as f:
             self.model = pickle.load(f)
-        with open("/tmp/iris_labels.json") as f:
+        with open(LABEL_PATH) as f:
             self.label_list = json.load(f)
 
     def __call__(self, flask_request):

python/ray/serve/examples/doc/tutorial_tensorflow.py

@@ -3,13 +3,14 @@
 from ray import serve
 
 import os
+import tempfile
 import numpy as np
 import requests
 # __doc_import_end__
 # yapf: enable
 
 # __doc_train_model_begin__
-TRAINED_MODEL_PATH = "/tmp/mnist_model.h5"
+TRAINED_MODEL_PATH = os.path.join(tempfile.gettempdir(), "mnist_model.h5")
 
 
 def train_and_save_model():
@@ -69,7 +70,7 @@ class TFMnistModel:
 
 # __doc_deploy_begin__
 client = serve.start()
-client.create_backend("tf:v1", TFMnistModel, "/tmp/mnist_model.h5")
+client.create_backend("tf:v1", TFMnistModel, TRAINED_MODEL_PATH)
 client.create_endpoint("tf_classifier", backend="tf:v1", route="/mnist")
 # __doc_deploy_end__