[Docs] [Data] Update Batch Inference end-to-end guide (ray-project#36018

) Updates batch inference end-to-end guide per offline discussion: Abridges the guide to make it more consumable Now that we've decided to keep the guide in the Ray Data docs, removes some of the duplicated information Cleans up configuration section Link out to the appropriate examples Adds more information on competitive analysis and why you should use Ray Data for batch inference. --------- Signed-off-by: amogkam <[email protected]>
Myasuka · Jun 6, 2023 · a66a0f2 · a66a0f2
1 parent 68ec6f8
commit a66a0f2
Show file tree

Hide file tree

Showing 8 changed files with 381 additions and 1,068 deletions.
diff --git a/.buildkite/pipeline.gpu_large.yml b/.buildkite/pipeline.gpu_large.yml
@@ -66,6 +66,10 @@
     - pip install -Ur ./python/requirements/ml/requirements_ml_docker.txt
     - ./ci/env/install-horovod.sh
     - ./ci/env/env_info.sh
+    # Test examples with newer version of `transformers`
+    # TODO(amogkam): Remove when https://github.com/ray-project/ray/issues/36011 
+    # is resolved.
+    - pip install -U transformers
     - bazel test --config=ci $(./scripts/bazel_export_options)
       --test_tag_filters=doctest,-cpu python/ray/... doc/...
 

diff --git a/doc/BUILD b/doc/BUILD
@@ -246,7 +246,6 @@ py_test_run_all_subdirectory(
     exclude = [
         "source/ray-air/doc_code/predictors.py",
         "source/data/doc_code/loading_data_untested.py",
-        "source/data/doc_code/torch_image_batch_trained.py"
     ],
     extra_srcs = [],
     tags = ["exclusive", "team:data"],
@@ -317,8 +316,18 @@ doctest(
             "source/ray-observability/user-guides/cli-sdk.rst",
             "source/rllib/rllib-env.rst",
             "source/rllib/rllib-sample-collection.rst",
+            "source/data/batch_inference.rst"
         ]
     ),
     size = "large",
     tags = ["team:none"]
 )
+
+
+doctest(
+    files = glob(
+        include=["source/data/batch_inference.rst"],
+    ),
+    tags = ["team:data"],
+    gpu = True
+)
diff --git a/doc/source/data/batch_inference.rst b/doc/source/data/batch_inference.rst
diff --git a/doc/source/data/doc_code/hf_quick_start.py b/doc/source/data/doc_code/hf_quick_start.py
diff --git a/doc/source/data/doc_code/pytorch_quick_start.py b/doc/source/data/doc_code/pytorch_quick_start.py
diff --git a/doc/source/data/doc_code/tf_quick_start.py b/doc/source/data/doc_code/tf_quick_start.py
diff --git a/doc/source/data/doc_code/torch_image_batch_trained.py b/doc/source/data/doc_code/torch_image_batch_trained.py
diff --git a/doc/source/data/user-guide.rst b/doc/source/data/user-guide.rst
@@ -15,7 +15,7 @@ show you how achieve several tasks.
     loading-data
     transforming-data
     consuming-data
-    batch_inference
     working-with-tensors
     data-internals
     performance-tips
+    batch_inference