You are viewing a plain text version of this content. The canonical link for it is here.
Posted to github@beam.apache.org by GitBox <gi...@apache.org> on 2022/07/12 19:42:26 UTC
[GitHub] [beam] AnandInguva commented on a diff in pull request #22088: sklearn runinference regression example
AnandInguva commented on code in PR #22088:
URL: https://github.com/apache/beam/pull/22088#discussion_r919245770
##########
sdks/python/apache_beam/examples/inference/sklearn_japanese_housing_regression.py:
##########
@@ -0,0 +1,164 @@
+#
+# Licensed to the Apache Software Foundation (ASF) under one or more
+# contributor license agreements. See the NOTICE file distributed with
+# this work for additional information regarding copyright ownership.
+# The ASF licenses this file to You under the Apache License, Version 2.0
+# (the "License"); you may not use this file except in compliance with
+# the License. You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# 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.
+#
+
+"""A pipeline that uses RunInference API on a regression about housing prices.
+
+This example uses the japanese housing data from kaggle.
+https://www.kaggle.com/datasets/nishiodens/japan-real-estate-transaction-prices
+
+Since the data has missing fields, this example illustrates how to split
+data and assign it to the models that are trained on different subsets of
+features. The predictions are then recombined.
+
+In order to set this example up, you will need two things.
+1. Build models (or use ours) and reference those via the model directory.
+2. Download the data from kaggle and host it.
+"""
+
+import argparse
+from typing import Iterable
+
+import apache_beam as beam
+from apache_beam.io.filesystems import FileSystems
+from apache_beam.ml.inference.base import RunInference
+from apache_beam.ml.inference.sklearn_inference import ModelFileType
+from apache_beam.ml.inference.sklearn_inference import SklearnModelHandlerPandas
+from apache_beam.options.pipeline_options import PipelineOptions
+from apache_beam.options.pipeline_options import SetupOptions
+import pandas
+
+MODELS = [{
+ 'name': 'all_features',
+ 'required_features': [
+ 'Area',
+ 'Year',
+ 'MinTimeToNearestStation',
+ 'MaxTimeToNearestStation',
+ 'TotalFloorArea',
+ 'Frontage',
+ 'Breadth',
+ 'BuildingYear'
+ ]
+},
+ {
+ 'name': 'floor_area',
+ 'required_features': ['Area', 'Year', 'TotalFloorArea']
+ },
+ {
+ 'name': 'stations',
+ 'required_features': [
+ 'Area',
+ 'Year',
+ 'MinTimeToNearestStation',
+ 'MaxTimeToNearestStation'
+ ]
+ }, {
+ 'name': 'no_features', 'required_features': ['Area', 'Year']
+ }]
+
+
+def sort_by_features(dataframe, max_size):
+ """ Partitions the dataframe by what data it has available."""
+ for i, model in enumerate(MODELS):
+ required_features = dataframe[model['required_features']]
+ if not required_features.isnull().any().any():
+ return i
+ return -1
+
+
+class LoadDataframe(beam.DoFn):
+ def process(self, file_name: str) -> Iterable[pandas.DataFrame]:
+ """ Loads data files as a pandas dataframe."""
+ file = FileSystems.open(file_name, 'rb')
+ dataframe = pandas.read_csv(file)
+ for i in range(dataframe.shape[0]):
+ yield dataframe.iloc[[i]]
+
+
+def report_predictions(prediction_result):
+ true_result = prediction_result.example['TradePrice'].values[0]
+ inference = prediction_result.inference
+ return 'True Price %.1f, Predicted Price %f' % (true_result, inference)
+
+
+def parse_known_args(argv):
+ """Parses args for the workflow."""
+ parser = argparse.ArgumentParser()
+ parser.add_argument(
+ '--input',
Review Comment:
```suggestion
'--input',
'--input_file'
```
```suggestion
'--input',
```
##########
sdks/python/apache_beam/examples/inference/sklearn_japanese_housing_regression.py:
##########
@@ -0,0 +1,164 @@
+#
+# Licensed to the Apache Software Foundation (ASF) under one or more
+# contributor license agreements. See the NOTICE file distributed with
+# this work for additional information regarding copyright ownership.
+# The ASF licenses this file to You under the Apache License, Version 2.0
+# (the "License"); you may not use this file except in compliance with
+# the License. You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# 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.
+#
+
+"""A pipeline that uses RunInference API on a regression about housing prices.
+
+This example uses the japanese housing data from kaggle.
+https://www.kaggle.com/datasets/nishiodens/japan-real-estate-transaction-prices
+
+Since the data has missing fields, this example illustrates how to split
+data and assign it to the models that are trained on different subsets of
+features. The predictions are then recombined.
+
+In order to set this example up, you will need two things.
+1. Build models (or use ours) and reference those via the model directory.
Review Comment:
Since we trained these models, we can host them on a bucket publicly available and tell the users to download them.
##########
sdks/python/apache_beam/examples/inference/sklearn_japanese_housing_regression.py:
##########
@@ -0,0 +1,164 @@
+#
+# Licensed to the Apache Software Foundation (ASF) under one or more
+# contributor license agreements. See the NOTICE file distributed with
+# this work for additional information regarding copyright ownership.
+# The ASF licenses this file to You under the Apache License, Version 2.0
+# (the "License"); you may not use this file except in compliance with
+# the License. You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# 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.
+#
+
+"""A pipeline that uses RunInference API on a regression about housing prices.
+
+This example uses the japanese housing data from kaggle.
+https://www.kaggle.com/datasets/nishiodens/japan-real-estate-transaction-prices
+
+Since the data has missing fields, this example illustrates how to split
+data and assign it to the models that are trained on different subsets of
+features. The predictions are then recombined.
+
+In order to set this example up, you will need two things.
+1. Build models (or use ours) and reference those via the model directory.
+2. Download the data from kaggle and host it.
+"""
+
+import argparse
+from typing import Iterable
+
+import apache_beam as beam
+from apache_beam.io.filesystems import FileSystems
+from apache_beam.ml.inference.base import RunInference
+from apache_beam.ml.inference.sklearn_inference import ModelFileType
+from apache_beam.ml.inference.sklearn_inference import SklearnModelHandlerPandas
+from apache_beam.options.pipeline_options import PipelineOptions
+from apache_beam.options.pipeline_options import SetupOptions
+import pandas
+
+MODELS = [{
+ 'name': 'all_features',
+ 'required_features': [
+ 'Area',
+ 'Year',
+ 'MinTimeToNearestStation',
+ 'MaxTimeToNearestStation',
+ 'TotalFloorArea',
+ 'Frontage',
+ 'Breadth',
+ 'BuildingYear'
+ ]
+},
+ {
+ 'name': 'floor_area',
+ 'required_features': ['Area', 'Year', 'TotalFloorArea']
+ },
+ {
+ 'name': 'stations',
+ 'required_features': [
+ 'Area',
+ 'Year',
+ 'MinTimeToNearestStation',
+ 'MaxTimeToNearestStation'
+ ]
+ }, {
+ 'name': 'no_features', 'required_features': ['Area', 'Year']
+ }]
+
+
+def sort_by_features(dataframe, max_size):
+ """ Partitions the dataframe by what data it has available."""
+ for i, model in enumerate(MODELS):
+ required_features = dataframe[model['required_features']]
+ if not required_features.isnull().any().any():
+ return i
+ return -1
+
+
+class LoadDataframe(beam.DoFn):
+ def process(self, file_name: str) -> Iterable[pandas.DataFrame]:
+ """ Loads data files as a pandas dataframe."""
+ file = FileSystems.open(file_name, 'rb')
+ dataframe = pandas.read_csv(file)
+ for i in range(dataframe.shape[0]):
+ yield dataframe.iloc[[i]]
+
+
+def report_predictions(prediction_result):
+ true_result = prediction_result.example['TradePrice'].values[0]
+ inference = prediction_result.inference
+ return 'True Price %.1f, Predicted Price %f' % (true_result, inference)
+
+
+def parse_known_args(argv):
+ """Parses args for the workflow."""
+ parser = argparse.ArgumentParser()
+ parser.add_argument(
+ '--input',
+ dest='input',
+ required=True,
+ help='A metadata file with all models, with references to models and '
+ 'information about all the files and data.')
+ parser.add_argument(
+ '--model_path',
+ dest='model_path',
Review Comment:
```suggestion
dest='models_path',
```
##########
sdks/python/apache_beam/examples/inference/sklearn_japanese_housing_regression.py:
##########
@@ -0,0 +1,164 @@
+#
+# Licensed to the Apache Software Foundation (ASF) under one or more
+# contributor license agreements. See the NOTICE file distributed with
+# this work for additional information regarding copyright ownership.
+# The ASF licenses this file to You under the Apache License, Version 2.0
+# (the "License"); you may not use this file except in compliance with
+# the License. You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# 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.
+#
+
+"""A pipeline that uses RunInference API on a regression about housing prices.
+
+This example uses the japanese housing data from kaggle.
+https://www.kaggle.com/datasets/nishiodens/japan-real-estate-transaction-prices
+
+Since the data has missing fields, this example illustrates how to split
+data and assign it to the models that are trained on different subsets of
+features. The predictions are then recombined.
+
+In order to set this example up, you will need two things.
+1. Build models (or use ours) and reference those via the model directory.
+2. Download the data from kaggle and host it.
+"""
+
+import argparse
+from typing import Iterable
+
+import apache_beam as beam
+from apache_beam.io.filesystems import FileSystems
+from apache_beam.ml.inference.base import RunInference
+from apache_beam.ml.inference.sklearn_inference import ModelFileType
+from apache_beam.ml.inference.sklearn_inference import SklearnModelHandlerPandas
+from apache_beam.options.pipeline_options import PipelineOptions
+from apache_beam.options.pipeline_options import SetupOptions
+import pandas
+
+MODELS = [{
+ 'name': 'all_features',
+ 'required_features': [
+ 'Area',
+ 'Year',
+ 'MinTimeToNearestStation',
+ 'MaxTimeToNearestStation',
+ 'TotalFloorArea',
+ 'Frontage',
+ 'Breadth',
+ 'BuildingYear'
+ ]
+},
+ {
+ 'name': 'floor_area',
+ 'required_features': ['Area', 'Year', 'TotalFloorArea']
+ },
+ {
+ 'name': 'stations',
+ 'required_features': [
+ 'Area',
+ 'Year',
+ 'MinTimeToNearestStation',
+ 'MaxTimeToNearestStation'
+ ]
+ }, {
+ 'name': 'no_features', 'required_features': ['Area', 'Year']
+ }]
+
+
+def sort_by_features(dataframe, max_size):
+ """ Partitions the dataframe by what data it has available."""
+ for i, model in enumerate(MODELS):
+ required_features = dataframe[model['required_features']]
+ if not required_features.isnull().any().any():
+ return i
+ return -1
+
+
+class LoadDataframe(beam.DoFn):
+ def process(self, file_name: str) -> Iterable[pandas.DataFrame]:
+ """ Loads data files as a pandas dataframe."""
+ file = FileSystems.open(file_name, 'rb')
+ dataframe = pandas.read_csv(file)
+ for i in range(dataframe.shape[0]):
+ yield dataframe.iloc[[i]]
+
+
+def report_predictions(prediction_result):
+ true_result = prediction_result.example['TradePrice'].values[0]
+ inference = prediction_result.inference
+ return 'True Price %.1f, Predicted Price %f' % (true_result, inference)
+
+
+def parse_known_args(argv):
+ """Parses args for the workflow."""
+ parser = argparse.ArgumentParser()
+ parser.add_argument(
+ '--input',
+ dest='input',
+ required=True,
+ help='A metadata file with all models, with references to models and '
+ 'information about all the files and data.')
+ parser.add_argument(
+ '--model_path',
+ dest='model_path',
Review Comment:
optional change. since the model_path has multiple models present
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@beam.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org