Spark NLP 6.0.3

CHANGELOG

@@ -1,5 +1,20 @@
 =======
-6.0.1
+6.0.3
+=======
+----------------
+New Features & Enhancements
+----------------
+* Introducing E5-V Universal Embeddings (SPARKNLP-1143)
+* Enhanced Chunking Strategies (SPARKNLP-1125)
+* New XML Reader (SPARKNLP-1119)
+
+----------------
+Bug Fixes
+----------------
+* Fixed typo for Excel reader notebook
+
+=======
+6.0.2
 =======
 ----------------
 New Features & Enhancements

README.md

@@ -63,7 +63,7 @@ $ java -version
 $ conda create -n sparknlp python=3.7 -y
 $ conda activate sparknlp
 # spark-nlp by default is based on pyspark 3.x
-$ pip install spark-nlp==6.0.2 pyspark==3.3.1
+$ pip install spark-nlp==6.0.3 pyspark==3.3.1
 ```
 
 In Python console or Jupyter `Python3` kernel:
@@ -129,7 +129,7 @@ For a quick example of using pipelines and models take a look at our official [d
 
 ### Apache Spark Support
 
-Spark NLP *6.0.2* has been built on top of Apache Spark 3.4 while fully supports Apache Spark 3.0.x, 3.1.x, 3.2.x, 3.3.x, 3.4.x, and 3.5.x
+Spark NLP *6.0.3* has been built on top of Apache Spark 3.4 while fully supports Apache Spark 3.0.x, 3.1.x, 3.2.x, 3.3.x, 3.4.x, and 3.5.x
 
 | Spark NLP | Apache Spark 3.5.x | Apache Spark 3.4.x | Apache Spark 3.3.x | Apache Spark 3.2.x | Apache Spark 3.1.x | Apache Spark 3.0.x | Apache Spark 2.4.x | Apache Spark 2.3.x |
 |-----------|--------------------|--------------------|--------------------|--------------------|--------------------|--------------------|--------------------|--------------------|
@@ -159,7 +159,7 @@ Find out more about 4.x `SparkNLP` versions in our official [documentation](http
 
 ### Databricks Support
 
-Spark NLP 6.0.2 has been tested and is compatible with the following runtimes:
+Spark NLP 6.0.3 has been tested and is compatible with the following runtimes:
 
 | **CPU**            | **GPU**            |
 |--------------------|--------------------|
@@ -176,7 +176,7 @@ We are compatible with older runtimes. For a full list check databricks support
 
 ### EMR Support
 
-Spark NLP 6.0.2 has been tested and is compatible with the following EMR releases:
+Spark NLP 6.0.3 has been tested and is compatible with the following EMR releases:
 
 | **EMR Release**    |
 |--------------------|

build.sbt

@@ -6,7 +6,7 @@ name := getPackageName(is_silicon, is_gpu, is_aarch64)
 
 organization := "com.johnsnowlabs.nlp"
 
-version := "6.0.2"
+version := "6.0.3"
 
 (ThisBuild / scalaVersion) := scalaVer
 

conda/meta.yaml

@@ -1,13 +1,13 @@
 {% set name = "spark-nlp" %}
-{% set version = "6.0.2" %}
+{% set version = "6.0.3" %}
 
 package:
   name: {{ name|lower }}
   version: {{ version }}
 
 source:
   url: https://pypi.io/packages/source/{{ name[0] }}/{{ name }}/spark_nlp-{{ version }}.tar.gz
-  sha256: 8b97358206809a123076bcd58aba3b6487086c95c6370be6a9a34f0d5568b43d
+  sha256: ff09f27c512401cff1ec3af572069b2e2af35b87a0f6737c5340538bac10faf7
 
 build:
   noarch: python

docs/en/transformer_entries/E5VEmbeddings.md

@@ -0,0 +1,133 @@
+{%- capture title -%}
+E5VEmbeddings
+{%- endcapture -%}
+
+{%- capture description -%}
+Universal multimodal embeddings using E5-V.
+
+E5-V is a multimodal embedding model that bridges the modality gap between text and images, enabling strong performance in cross-modal retrieval, classification, clustering, and more. It supports both image+text and text-only embedding scenarios, and is fine-tuned from lmms-lab/llama3-llava-next-8b. The default model is `"e5v_int4"`.
+
+Note that this annotator is only supported for Spark Versions 3.4 and up.
+
+Pretrained models can be loaded with `pretrained` of the companion object:
+
+```scala
+val embeddings = E5VEmbeddings.pretrained()
+  .setInputCols("image_assembler")
+  .setOutputCol("e5v")
+```
+
+For available pretrained models please see the
+[Models Hub](https://sparknlp.org/models?q=E5V).
+
+For extended examples of usage, see
+[E5VEmbeddingsTestSpec](https://github.com/JohnSnowLabs/spark-nlp/blob/master/src/test/scala/com/johnsnowlabs/nlp/embeddings/E5VEmbeddingsTestSpec.scala).
+
+**Sources** :
+
+- [E5-V: Universal Embeddings with Multimodal Large Language Models (arXiv)](https://arxiv.org/abs/2407.12580)
+- [Hugging Face Model Card](https://huggingface.co/royokong/e5-v)
+- [E5-V Github Repository](https://github.com/kongds/E5-V)
+{%- endcapture -%}
+
+{%- capture input_anno -%}
+IMAGE
+{%- endcapture -%}
+
+{%- capture output_anno -%}
+SENTENCE_EMBEDDINGS
+{%- endcapture -%}
+
+{%- capture python_example -%}
+# Image + Text Embedding
+import sparknlp
+from sparknlp.base import *
+from sparknlp.annotator import *
+from pyspark.ml import Pipeline
+from pyspark.sql.functions import lit
+
+image_df = spark.read.format("image").option("dropInvalid", True).load(imageFolder)
+imagePrompt = "<|start_header_id|>user<|end_header_id|>\n\n<image>\\nSummary above image in one word: <|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n \n"
+test_df = image_df.withColumn("text", lit(imagePrompt))
+imageAssembler = ImageAssembler() \
+    .setInputCol("image") \
+    .setOutputCol("image_assembler")
+e5vEmbeddings = E5VEmbeddings.pretrained() \
+    .setInputCols(["image_assembler"]) \
+    .setOutputCol("e5v")
+pipeline = Pipeline().setStages([
+    imageAssembler,
+    e5vEmbeddings
+])
+result = pipeline.fit(test_df).transform(test_df)
+result.select("e5v.embeddings").show(truncate=False)
+
+# Text-Only Embedding
+from sparknlp.util import EmbeddingsDataFrameUtils
+textPrompt = "<|start_header_id|>user<|end_header_id|>\n\n<sent>\\nSummary above sentence in one word: <|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n \n"
+textDesc = "A cat sitting in a box."
+nullImageDF = spark.createDataFrame(
+    spark.sparkContext.parallelize([EmbeddingsDataFrameUtils.emptyImageRow]),
+    EmbeddingsDataFrameUtils.imageSchema)
+textDF = nullImageDF.withColumn("text", lit(textPrompt.replace("<sent>", textDesc)))
+e5vEmbeddings = E5VEmbeddings.pretrained() \
+    .setInputCols(["image"]) \
+    .setOutputCol("e5v")
+result = e5vEmbeddings.transform(textDF)
+result.select("e5v.embeddings").show(truncate=False)
+{%- endcapture -%}
+
+{%- capture scala_example -%}
+// Image + Text Embedding
+import org.apache.spark.sql.functions.lit
+import com.johnsnowlabs.nlp.base.ImageAssembler
+import com.johnsnowlabs.nlp.embeddings.E5VEmbeddings
+import org.apache.spark.ml.Pipeline
+
+val imageDF = spark.read.format("image").option("dropInvalid", value = true).load(imageFolder)
+val imagePrompt = "<|start_header_id|>user<|end_header_id|>\n\n<image>\\nSummary above image in one word: <|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n \n"
+val testDF = imageDF.withColumn("text", lit(imagePrompt))
+val imageAssembler = new ImageAssembler().setInputCol("image").setOutputCol("image_assembler")
+val e5vEmbeddings = E5VEmbeddings.pretrained()
+  .setInputCols("image_assembler")
+  .setOutputCol("e5v")
+val pipeline = new Pipeline().setStages(Array(imageAssembler, e5vEmbeddings))
+val result = pipeline.fit(testDF).transform(testDF)
+result.select("e5v.embeddings").show(truncate = false)
+
+// Text-Only Embedding
+import com.johnsnowlabs.nlp.util.EmbeddingsDataFrameUtils.{emptyImageRow, imageSchema}
+val textPrompt = "<|start_header_id|>user<|end_header_id|>\n\n<sent>\\nSummary above sentence in one word: <|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n \n"
+val textDesc = "A cat sitting in a box."
+val nullImageDF = spark.createDataFrame(spark.sparkContext.parallelize(Seq(emptyImageRow)), imageSchema)
+val textDF = nullImageDF.withColumn("text", lit(textPrompt.replace("<sent>", textDesc)))
+val e5vEmbeddings = E5VEmbeddings.pretrained()
+  .setInputCols("image")
+  .setOutputCol("e5v")
+val result2 = e5vEmbeddings.transform(textDF)
+result2.select("e5v.embeddings").show(truncate = false)
+{%- endcapture -%}
+
+{%- capture api_link -%}
+[E5VEmbeddings](/api/com/johnsnowlabs/nlp/embeddings/E5VEmbeddings)
+{%- endcapture -%}
+
+{%- capture python_api_link -%}
+[E5VEmbeddings](/api/python/reference/autosummary/sparknlp/annotator/cv/e5v_embeddings/index.html#sparknlp.annotator.cv.e5v_embeddings.E5VEmbeddings)
+{%- endcapture -%}
+
+{%- capture source_link -%}
+[E5VEmbeddings](https://github.com/JohnSnowLabs/spark-nlp/tree/master/src/main/scala/com/johnsnowlabs/nlp/embeddings/E5VEmbeddings.scala)
+{%- endcapture -%}
+
+{% include templates/anno_template.md
+  title=title
+  description=description
+  input_anno=input_anno
+  output_anno=output_anno
+  python_example=python_example
+  scala_example=scala_example
+  api_link=api_link
+  python_api_link=python_api_link
+  source_link=source_link
+%}