sparknlp.annotator.cv.mllama_for_multimodal
#
Module Contents#
Classes#
MLLamaForMultimodal can load LLAMA 3.2 Vision models for visual question answering. |
- class MLLamaForMultimodal(classname='com.johnsnowlabs.nlp.annotators.cv.MLLamaForMultimodal', java_model=None)[source]#
MLLamaForMultimodal can load LLAMA 3.2 Vision models for visual question answering. The model consists of a vision encoder, a text encoder, and a text decoder. The vision encoder encodes the input image, the text encoder processes the input question alongside the image encoding, and the text decoder generates the answer to the question.
The Llama 3.2-Vision collection comprises pretrained and instruction-tuned multimodal large language models (LLMs) available in 11B and 90B sizes. These models are optimized for visual recognition, image reasoning, captioning, and answering general questions about images. The models outperform many open-source and proprietary multimodal models on standard industry benchmarks.
Pretrained models can be loaded with
pretrained()
of the companion object:>>> visualQAClassifier = MLLamaForMultimodal.pretrained() \ ... .setInputCols(["image_assembler"]) \ ... .setOutputCol("answer")
The default model is βllama_3_2_11b_vision_instruct_int4β, if no name is provided.
For available pretrained models, refer to the Models Hub.
Models from the HuggingFace π€ Transformers library are also compatible with Spark NLP π. To check compatibility and learn how to import them, see Import Transformers into Spark NLP π. For extended examples, refer to the MLLamaForMultimodal Test Suite.
Input Annotation types
Output Annotation type
IMAGE
DOCUMENT
- Parameters:
- batchSizeint, optional
Batch size. Larger values allow faster processing but require more memory, by default 2.
- configProtoBytesbytes, optional
ConfigProto from TensorFlow, serialized into a byte array.
- maxSentenceLengthint, optional
Maximum sentence length to process, by default 50.
Examples
>>> import sparknlp >>> from sparknlp.base import * >>> from sparknlp.annotator import * >>> from pyspark.ml import Pipeline >>> from pyspark.sql.functions import lit >>> image_df = SparkSessionForTest.spark.read.format("image").load(path=images_path) >>> test_df = image_df.withColumn( ... "text", ... lit("<|begin_of_text|><|begin_of_text|><|start_header_id|>user<|end_header_id|>
<|image|>What is unusual on this image?<|eot_id|><|start_header_id|>assistant<|end_header_id|>
β) β¦ ) >>> imageAssembler = ImageAssembler() β¦ .setInputCol(βimageβ) β¦ .setOutputCol(βimage_assemblerβ) >>> visualQAClassifier = MLLamaForMultimodal.pretrained() β¦ .setInputCols(βimage_assemblerβ) β¦ .setOutputCol(βanswerβ) >>> pipeline = Pipeline().setStages([ β¦ imageAssembler, β¦ visualQAClassifier β¦ ]) >>> result = pipeline.fit(test_df).transform(test_df) >>> result.select(βimage_assembler.originβ, βanswer.resultβ).show(truncate=False) +βββββββββββββ+βββββββββββββββββββββββ-+ |origin |result | +--------------------------------------+----------------------------------------------------------------------+ |[file:///content/images/cat_image.jpg]|[The unusual aspect of this picture is the presence of two cats lying on a pink couch]| +βββββββββββββ+βββββββββββββββββββββββ-+
- setMaxSentenceSize(value)[source]#
Sets Maximum sentence length that the annotator will process, by default 50.
- Parameters:
- valueint
Maximum sentence length that the annotator will process
- setIgnoreTokenIds(value)[source]#
A list of token ids which are ignored in the decoderβs output.
- Parameters:
- valueList[int]
The words to be filtered out
- setConfigProtoBytes(b)[source]#
Sets configProto from tensorflow, serialized into byte array.
- Parameters:
- bList[int]
ConfigProto from tensorflow, serialized into byte array
- setMinOutputLength(value)[source]#
Sets minimum length of the sequence to be generated.
- Parameters:
- valueint
Minimum length of the sequence to be generated
- setMaxOutputLength(value)[source]#
Sets maximum length of output text.
- Parameters:
- valueint
Maximum length of output text
- setDoSample(value)[source]#
Sets whether or not to use sampling, use greedy decoding otherwise.
- Parameters:
- valuebool
Whether or not to use sampling; use greedy decoding otherwise
- setTemperature(value)[source]#
Sets the value used to module the next token probabilities.
- Parameters:
- valuefloat
The value used to module the next token probabilities
- setTopK(value)[source]#
Sets the number of highest probability vocabulary tokens to keep for top-k-filtering.
- Parameters:
- valueint
Number of highest probability vocabulary tokens to keep
- setTopP(value)[source]#
Sets the top cumulative probability for vocabulary tokens.
If set to float < 1, only the most probable tokens with probabilities that add up to
topP
or higher are kept for generation.- Parameters:
- valuefloat
Cumulative probability for vocabulary tokens
- setRepetitionPenalty(value)[source]#
Sets the parameter for repetition penalty. 1.0 means no penalty.
- Parameters:
- valuefloat
The repetition penalty
References
See Ctrl: A Conditional Transformer Language Model For Controllable Generation for more details.
- setNoRepeatNgramSize(value)[source]#
Sets size of n-grams that can only occur once.
If set to int > 0, all ngrams of that size can only occur once.
- Parameters:
- valueint
N-gram size can only occur once
- setBeamSize(value)[source]#
Sets the number of beam size for beam search, by default 4.
- Parameters:
- valueint
Number of beam size for beam search
- static loadSavedModel(folder, spark_session, use_openvino=False)[source]#
Loads a locally saved model.
- Parameters:
- folderstr
Folder of the saved model
- spark_sessionpyspark.sql.SparkSession
The current SparkSession
- Returns:
- CLIPForZeroShotClassification
The restored model
- static pretrained(name='llama_3_2_11b_vision_instruct_int4', lang='en', remote_loc=None)[source]#
Downloads and loads a pretrained model.
- Parameters:
- namestr, optional
Name of the pretrained model, by default βllama_3_2_11b_vision_instruct_int4β
- langstr, optional
Language of the pretrained model, by default βenβ
- remote_locstr, optional
Optional remote address of the resource, by default None. Will use Spark NLPs repositories otherwise.
- Returns:
- MLLamaForMultimodal
The restored model