Build, Scale, and Deploy Deep Learning Pipelines Using Apache Spark
•
2 recomendaciones•1,381 vistas
Deep learning has shown tremendous successes, yet it often requires a lot of effort to leverage its power. Existing deep learning frameworks require writing a lot of code to run a model, let alone in a distributed manner. Deep Learning Pipelines is an Apache Spark Package library that makes practical deep learning simple based on the Spark MLlib Pipelines API. Leveraging Spark, Deep Learning Pipelines scales out many compute-intensive deep learning tasks. In this talk, we discuss the philosophy behind Deep Learning Pipelines, as well as the main tools it provides, how they fit into the deep learning ecosystem, and how they demonstrate Spark's role in deep learning.
Recomendados
Recomendados
Más contenido relacionado
La actualidad más candente
La actualidad más candente (20)
Similar a Build, Scale, and Deploy Deep Learning Pipelines Using Apache Spark
Similar a Build, Scale, and Deploy Deep Learning Pipelines Using Apache Spark (20)
Más de Databricks
Más de Databricks (20)
Último
Último (20)
Build, Scale, and Deploy Deep Learning Pipelines Using Apache Spark
- 1. Build, Scale, and Deploy Deep Learning Pipelines Using Apache Spark Bay Area Spark Meetup Nov8, 2017 Sue Ann Hong, Databricks
- 2. This talk • Deep Learning at scale: current state • Deep Learning Pipelines: the philosophy • End-to-end workflow with DL Pipelines • Future
- 3. Deep Learning at Scale : current state 3put your #assignedhashtag here by setting the
- 4. What is Deep Learning? • A set of machine learning techniques that use layers that transform numerical inputs • Classification • Regression • Arbitrary mapping • Popular in the 80’s as Neural Networks • Recently came back thanks to advances in data collection, computation techniques, and hardware.
- 5. Success of Deep Learning Tremendous success for applications with complex data • AlphaGo • Image interpretation • Automatictranslation • Speech recognition
- 6. Deep Learning is often challenging Labeled Data Compute Resources & Time Engineer hours • Tedious or difficult to distribute computations • No exact science around deep learning à lots of tweaking • Low level APIs with steep learning curve • Complex models à need a lot of data
- 7. Deep Learning in industry • Currently limited adoption • Huge potential beyond the industrial giants • How do we accelerate the road to massive availability?
- 9. Deep Learning Pipelines: Deep Learning with Simplicity • Open-source Databricks library • Focuses on easeof useand integration • without sacrificing performance
- 10. Deep Learning is often challenging Compute Resources & Time Engineer hours Labeled Data • Tedious or difficult to distribute computations • No exact science around deep learning à lots of tweaking • Low level APIs with steep learning curve • Complex models à need a lot of data
- 11. Deep Learning is often challenging • Tedious or difficult to distribute computations • No exact science around deep learning à lots of tweaking • Low level APIs with steep learning curve • Complex models à need a lot of data
- 12. • Tedious or difficult to distribute computations • No exact science around deep learning à lots of tweaking • Low level APIs with steep learning curve • Complex models à need a lot of data Instead • Be easy to scale • Require little tweaking • Be easy to write • Require little or no data Common workflows should
- 13. How • Be easy to scale • Require little tweaking • Be easy to write • Require little or no data Common workflows should • Use Apache Spark for scaling out common tasks • Leverage well-knownmodel architectures • Integrate with MLlib Pipelines API to capture ML workflowconcisely • Leverage pre-trained models for common tasks Apache Spark for scaling out MLlib Pipelines API pre-trained models
- 14. Demo: Build a visual recommendation AI 10minutes 7lines of code Elastic Scale-out using Apache Spark MLlib Pipelines API Leverage pre-trained models 0labels
- 16. To the notebook!
- 18. End-to-EndWorkflow with Deep Learning Pipelines 18put your #assignedhashtag here by setting the
- 19. A typical Deep Learning workflow • Load data (images, text, time series, …) • Interactive work • Train • Select an architecture for a neural network • Optimize the weights of the NN • Evaluateresults, potentially re-train • Apply: • Pass the data through the NN to produce new features or output Load data Interactive work Train Evaluate Apply
- 20. A typical Deep Learning workflow Load data Interactive work Train Evaluate Apply • Image loading in Spark • Distributed batch prediction • Deploying models in SQL • Transfer learning • Distributed tuning • Pre-trained models
- 21. Deep Learning Pipelines • Load data • Interactive work • Train • Evaluate model • Apply
- 22. Adds support for images in Spark • ImageSchema, reader, conversion functions to/from numpy arrays • Most of the tools we’ll describe work on ImageSchema columns from sparkdl import readImages image_df = readImages(sample_img_dir)
- 23. Upcoming: built-in support in Spark • Spark-21866 • Contributing image format & reading to Spark • Targeted for Spark 2.3 • Joint work with Microsoft 23put your #assignedhashtag here by setting the
- 24. Deep Learning Pipelines • Load data • Interactive work • Train • Evaluate model • Apply
- 25. Applying popular models • Popular pre-trained models accessible through MLlib Transformers predictor = DeepImagePredictor(inputCol="image", outputCol="predicted_labels", modelName="InceptionV3") predictions_df = predictor.transform(image_df)
- 26. Applying popular models predictor = DeepImagePredictor(inputCol="image", outputCol="predicted_labels", modelName="InceptionV3") predictions_df = predictor.transform(image_df)
- 27. Deep Learning Pipelines • Load data • Interactive work • Train • Evaluate model • Apply Hyperparameter tuning Transfer learning
- 28. Deep Learning Pipelines • Load data • Interactive work • Train • Evaluate model • Apply Hyperparameter tuning Transfer learning
- 29. Transfer learning • Pre-trained models may not be directly applicable • New domain, e.g. shoes • Training from scratch requires • Enormous amounts of data • A lot of compute resources & time • Idea: intermediate representations learned for one task may be useful for other related tasks
- 30. Transfer Learning SoftMax GIANT PANDA 0.9 RACCOON 0.05 RED PANDA 0.01 …
- 34. Transfer Learning as a Pipeline DeepImageFeaturizer Image Loading Preprocessing Logistic Regression MLlib Pipeline
- 35. Transfer Learning as a Pipeline 35put your #assignedhashtag here by setting the featurizer = DeepImageFeaturizer(inputCol="image", outputCol="features", modelName="InceptionV3") lr = LogisticRegression(labelCol="label") p = Pipeline(stages=[featurizer, lr]) p_model = p.fit(train_df)
- 36. Transfer Learning • Usually for classification tasks • Similar task, new domain • But other forms of learning leveraging learned representations can be loosely considered transfer learning
- 38. Featurization for similarity-based ML DeepImageFeaturizer Image Loading Preprocessing Logistic Regression
- 39. Featurization for similarity-based ML DeepImageFeaturizer Image Loading Preprocessing Clustering KMeans GaussianMixture Nearest Neighbor KNN LSH Distance computation
- 40. Featurization for similarity-based ML DeepImageFeaturizer Image Loading Preprocessing Clustering KMeans GaussianMixture Nearest Neighbor KNN LSH Distance computation Duplicate Detection Recommendation Anomaly Detection Search result diversification
- 41. Deep Learning Pipelines • Load data • Interactive work • Train • Evaluate model • Apply Hyperparameter tuning Transfer learning
- 42. Deep Learning Pipelines • Load data • Interactive work • Train • Evaluate model • Apply
- 43. Deep Learning Pipelines • Load data • Interactive work • Train • Evaluate model • Apply Spark SQL Batch prediction s
- 44. Deep Learning Pipelines • Load data • Interactive work • Train • Evaluate model • Apply Spark SQL Batch prediction
- 45. Batch prediction as an MLlib Transformer • A model is a Transformer in MLlib • DataFrame goes in, DataFrame comes out with output columns predictor = XXTransformer(inputCol="image", outputCol="prediction", modelSpecification={…}) predictions = predictor.transform(test_df)
- 46. Hierarchy of DL transformers for images 46 TFImageTransformer KerasImageTransformer DeepImageFeaturizerDeepImagePredictor NamedImageTransformer(model_name) (keras.Model) (tf.Graph) Input (Image) Output (vector) model_namemodel_name Keras.Model tf.Graph
- 47. Hierarchy of DL transformers for images 47 TFImageTransformer KerasImageTransformer DeepImageFeaturizerDeepImagePredictor NamedImageTransformer(model_name) (keras.Model) (tf.Graph) Input (Image) Output (vector) model_namemodel_name Keras.Model tf.Graph
- 48. Hierarchy of DL transformers 48 TFImageTransformer KerasImageTransformer DeepImageF eaturizer DeepImageP redictor NamedImageTransformer ) Input (Image ) Output (vector ) model_namemodel_name Keras.Model tf.Graph TFTextTransformer KerasTextTransformer DeepTextFeat urizer DeepTextPre dictor NamedTextTransformerInput (Text) Output (vector) model_namemodel_name Keras.Model tf.Graph TFTransformer
- 49. Hierarchy of DL transformers 49 TFImageTransformer KerasImageTransformer DeepImageFeaturizerDeepImagePredictor NamedImageTransformer TFTextTransformer KerasTextTransformer DeepTextFeaturizerDeepTextPredictor NamedTextTransformer TFTransformer
- 50. Hierarchy of DL transformers 50 TFImageTransformer KerasImageTransformer DeepImageFeaturizerDeepImagePredictor NamedImageTransformer TFTextTransformer KerasTextTransformer DeepTextFeaturizerDeepTextPredictor NamedTextTransformer TFTransformer
- 51. CommonTasks Easy Advanced Ones Possible 51 TFImageTransformer KerasImageTransformer DeepImageFeaturizerDeepImagePredictor NamedImageTransformer TFTextTransformer KerasTextTransformer DeepTextFeaturizerDeepTextPredictor NamedTextTransformer TFTransformer
- 52. CommonTasks Easy Advanced Ones Possible 52 TFImageTransformer KerasImageTransformer DeepImageFeaturizer DeepImagePredictor NamedImageTransformer TFTextTransformer KerasTextTransformer DeepTextFeaturizer DeepTextPredictor NamedTextTransformer TFTransformer Pre-built Solutions 80%-built Solutions Self-built Solutions
- 53. Deep Learning Pipelines : Future In progress • Text featurization (embeddings) • TFTransformer for arbitrary vectors Future directions • Non-image data domains: video, text, speech, … • Distributed training • Support for more backends, e.g. MXNet, PyTorch, BigDL
- 55. Resources DL Pipelines GitHub Repo, Spark Summit Europe 2017 Deep Dive Blog posts & webinars (http://databricks.com/blog) • Deep Learning Pipelines • GPU acceleration in Databricks • BigDL on Databricks • Deep Learning and Apache Spark Docs for Deep Learningon Databricks (http://docs.databricks.com) • Getting started • Deep Learning Pipelines Example • Spark integration