strickvl's picture
Add new SentenceTransformer model.
c54a34b verified
metadata
base_model: Snowflake/snowflake-arctic-embed-m
datasets: []
language:
  - en
library_name: sentence-transformers
license: apache-2.0
metrics:
  - cosine_accuracy@1
  - cosine_accuracy@3
  - cosine_accuracy@5
  - cosine_accuracy@10
  - cosine_precision@1
  - cosine_precision@3
  - cosine_precision@5
  - cosine_precision@10
  - cosine_recall@1
  - cosine_recall@3
  - cosine_recall@5
  - cosine_recall@10
  - cosine_ndcg@10
  - cosine_mrr@10
  - cosine_map@100
pipeline_tag: sentence-similarity
tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:1490
  - loss:MatryoshkaLoss
  - loss:TripletLoss
widget:
  - source_sentence: >-
      Where is the global configuration directory located in ZenML's default
      setup?
    sentences:
      - >-
        'default' ...


        Creating default user 'default' ...Creating default stack for user
        'default' in workspace default...


        Active workspace not set. Setting it to the default.


        The active stack is not set. Setting the active stack to the default
        workspace stack.


        Using the default store for the global config.


        Unable to find ZenML repository in your current working directory
        (/tmp/folder) or any parent directories. If you want to use an existing
        repository which is in a different location, set the environment
        variable 'ZENML_REPOSITORY_PATH'. If you want to create a new
        repository, run zenml init.


        Running without an active repository root.


        Using the default local database.


        Running with active workspace: 'default' (global)


        ┏━━━━━━━━┯━━━━━━━━━━━━┯━━━━━━━━┯━━━━━━━━━┯━━━━━━━━━━━━━━━━┯━━━━━━━━━━━━━━┓


         ACTIVE  STACK NAME  SHARED  OWNER    ARTIFACT_STORE  ORCHESTRATOR
        


        ┠────────┼────────────┼────────┼─────────┼────────────────┼──────────────┨


           👉    default           default  default         default     
        


        ┗━━━━━━━━┷━━━━━━━━━━━━┷━━━━━━━━┷━━━━━━━━━┷━━━━━━━━━━━━━━━━┷━━━━━━━━━━━━━━┛


        The following is an example of the layout of the global config directory
        immediately after initialization:


        /home/stefan/.config/zenml   <- Global Config Directory


        ├── config.yaml              <- Global Configuration Settings


        └── local_stores             <- Every Stack component that stores
        information


        |                           locally will have its own subdirectory here.


        ├── a1a0d3d0-d552-4a80-be09-67e5e29be8ee   <- e.g. Local Store path for
        the


        |                                             `default` local Artifact
        Store


        └── default_zen_store


        └── zenml.db         <- SQLite database where ZenML data (stacks,


        components, etc) are stored by default.


        As shown above, the global config directory stores the following
        information:
      - How do you configure the network settings on a Linux server?
      - >-
        Reranking for better retrieval


        Add reranking to your RAG inference for better retrieval performance.


        Rerankers are a crucial component of retrieval systems that use LLMs.
        They help improve the quality of the retrieved documents by reordering
        them based on additional features or scores. In this section, we'll
        explore how to add a reranker to your RAG inference pipeline in ZenML.


        In previous sections, we set up the overall workflow, from data
        ingestion and preprocessing to embeddings generation and retrieval. We
        then set up some basic evaluation metrics to assess the performance of
        our retrieval system. A reranker is a way to squeeze a bit of extra
        performance out of the system by reordering the retrieved documents
        based on additional features or scores.


        As you can see, reranking is an optional addition we make to what we've
        already set up. It's not strictly necessary, but it can help improve the
        relevance and quality of the retrieved documents, which in turn can lead
        to better responses from the LLM. Let's dive in!


        PreviousEvaluation in practice


        NextUnderstanding reranking


        Last updated 1 month ago
  - source_sentence: >-
      Where can I find the instructions to enable CUDA for GPU-backed hardware
      in ZenML SDK Docs?
    sentences:
      - >-
        Migration guide 0.39.1 → 0.41.0


        How to migrate your ZenML pipelines and steps from version <=0.39.1 to
        0.41.0.


        ZenML versions 0.40.0 to 0.41.0 introduced a new and more flexible
        syntax to define ZenML steps and pipelines. This page contains code
        samples that show you how to upgrade your steps and pipelines to the new
        syntax.


        Newer versions of ZenML still work with pipelines and steps defined
        using the old syntax, but the old syntax is deprecated and will be
        removed in the future.


        Overview


        from typing import Optional


        from zenml.steps import BaseParameters, Output, StepContext, step


        from zenml.pipelines import pipeline


        # Define a Step


        class MyStepParameters(BaseParameters):


        param_1: int


        param_2: Optional[float] = None


        @step


        def my_step(


        params: MyStepParameters, context: StepContext,


        ) -> Output(int_output=int, str_output=str):


        result = int(params.param_1 * (params.param_2 or 1))


        result_uri = context.get_output_artifact_uri()


        return result, result_uri


        # Run the Step separately


        my_step.entrypoint()


        # Define a Pipeline


        @pipeline


        def my_pipeline(my_step):


        my_step()


        step_instance = my_step(params=MyStepParameters(param_1=17))


        pipeline_instance = my_pipeline(my_step=step_instance)


        # Configure and run the Pipeline


        pipeline_instance.configure(enable_cache=False)


        schedule = Schedule(...)


        pipeline_instance.run(schedule=schedule)


        # Fetch the Pipeline Run


        last_run = pipeline_instance.get_runs()[0]


        int_output = last_run.get_step["my_step"].outputs["int_output"].read()


        from typing import Annotated, Optional, Tuple


        from zenml import get_step_context, pipeline, step


        from zenml.client import Client


        # Define a Step


        @step


        def my_step(


        param_1: int, param_2: Optional[float] = None


        ) -> Tuple[Annotated[int, "int_output"], Annotated[str, "str_output"]]:


        result = int(param_1 * (param_2 or 1))


        result_uri = get_step_context().get_output_artifact_uri()


        return result, result_uri


        # Run the Step separately


        my_step()


        # Define a Pipeline


        @pipeline
      - >-
        How do I integrate Google Cloud VertexAI into my existing Kubernetes
        cluster?
      - >2-
         SDK Docs .

        Enabling CUDA for GPU-backed hardwareNote that if you wish to use this
        step operator to run steps on a GPU, you will need to follow the
        instructions on this page to ensure that it works. It requires adding
        some extra settings customization and is essential to enable CUDA for
        the GPU to give its full acceleration.


        PreviousStep Operators


        NextGoogle Cloud VertexAI


        Last updated 19 days ago
  - source_sentence: >-
      What are the special metadata types supported by ZenML and how are they
      used?
    sentences:
      - >-
        Special Metadata Types


        Tracking your metadata.


        ZenML supports several special metadata types to capture specific kinds
        of information. Here are examples of how to use the special types Uri,
        Path, DType, and StorageSize:


        from zenml.metadata.metadata_types import StorageSize, DType


        from zenml import log_artifact_metadata


        log_artifact_metadata(


        metadata={


        "dataset_source": Uri("gs://my-bucket/datasets/source.csv"),


        "preprocessing_script": Path("/scripts/preprocess.py"),


        "column_types": {


        "age": DType("int"),


        "income": DType("float"),


        "score": DType("int")


        },


        "processed_data_size": StorageSize(2500000)


        In this example:


        Uri is used to indicate a dataset source URI.


        Path is used to specify the filesystem path to a preprocessing script.


        DType is used to describe the data types of specific columns.


        StorageSize is used to indicate the size of the processed data in bytes.


        These special types help standardize the format of metadata and ensure
        that it is logged in a consistent and interpretable manner.


        PreviousGroup metadata


        NextFetch metadata within steps


        Last updated 19 days ago
      - >-
        Configure a code repository


        Connect a Git repository to ZenML to track code changes and collaborate
        on MLOps projects.


        Throughout the lifecycle of a MLOps pipeline, it can get quite tiresome
        to always wait for a Docker build every time after running a pipeline
        (even if the local Docker cache is used). However, there is a way to
        just have one pipeline build and keep reusing it until a change to the
        pipeline environment is made: by connecting a code repository.


        With ZenML, connecting to a Git repository optimizes the Docker build
        processes. It also has the added bonus of being a better way of managing
        repository changes and enabling better code collaboration. Here is how
        the flow changes when running a pipeline:


        You trigger a pipeline run on your local machine. ZenML parses the
        @pipeline function to determine the necessary steps.


        The local client requests stack information from the ZenML server, which
        responds with the cloud stack configuration.


        The local client detects that we're using a code repository and requests
        the information from the git repo.


        Instead of building a new Docker image, the client checks if an existing
        image can be reused based on the current Git commit hash and other
        environment metadata.


        The client initiates a run in the orchestrator, which sets up the
        execution environment in the cloud, such as a VM.


        The orchestrator downloads the code directly from the Git repository and
        uses the existing Docker image to run the pipeline steps.


        Pipeline steps execute, storing artifacts in the cloud-based artifact
        store.


        Throughout the execution, the pipeline run status and metadata are
        reported back to the ZenML server.


        By connecting a Git repository, you avoid redundant builds and make your
        MLOps processes more efficient. Your team can work on the codebase
        simultaneously, with ZenML handling the version tracking and ensuring
        that the correct code version is always used for each run.


        Creating a GitHub Repository
      - >-
        Can you explain the process of setting up a virtual environment in
        Python?
  - source_sentence: >-
      What are the benefits of deploying stack components directly from the
      ZenML CLI?
    sentences:
      - >-
        ─────────────────────────────────────────────────┨┃ RESOURCE TYPES   │
        🔵 gcp-generic, 📦 gcs-bucket, 🌀 kubernetes-cluster, 🐳 docker-registry



        ┠──────────────────┼──────────────────────────────────────────────────────────────────────────┨


         RESOURCE NAME    
        <multiple>                                                              
        


        ┠──────────────────┼──────────────────────────────────────────────────────────────────────────┨


         SECRET ID        
        4694de65-997b-4929-8831-b49d5e067b97                                    
        


        ┠──────────────────┼──────────────────────────────────────────────────────────────────────────┨


         SESSION DURATION 
        N/A                                                                     
        


        ┠──────────────────┼──────────────────────────────────────────────────────────────────────────┨


         EXPIRES IN       
        59m46s                                                                  
        


        ┠──────────────────┼──────────────────────────────────────────────────────────────────────────┨


         OWNER            
        default                                                                 
        


        ┠──────────────────┼──────────────────────────────────────────────────────────────────────────┨


         WORKSPACE        
        default                                                                 
        


        ┠──────────────────┼──────────────────────────────────────────────────────────────────────────┨


         SHARED           
                                                                              
        


        ┠──────────────────┼──────────────────────────────────────────────────────────────────────────┨


         CREATED_AT        2023-05-19
        09:04:33.557126                                               


        ┠──────────────────┼──────────────────────────────────────────────────────────────────────────┨


         UPDATED_AT        2023-05-19
        09:04:33.557127                                               


        ┗━━━━━━━━━━━━━━━━━━┷━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┛


        Configuration


        ┏━━━━━━━━━━━━┯━━━━━━━━━━━━┓
      - How do you set up a custom service account for Vertex AI?
      - >-
        ⚒️Manage stacks


        Deploying your stack components directly from the ZenML CLI


        The first step in running your pipelines on remote infrastructure is to
        deploy all the components that you would need, like an MLflow tracking
        server, a Seldon Core model deployer, and more to your cloud.


        This can bring plenty of benefits like scalability, reliability, and
        collaboration. ZenML eases the path to production by providing a
        seamless way for all tools to interact with others through the use of
        abstractions. However, one of the most painful parts of this process,
        from what we see on our Slack and in general, is the deployment of these
        stack components.


        Deploying and managing MLOps tools is tricky 😭😵‍💫


        It is not trivial to set up all the different tools that you might need
        for your pipeline.


        🌈 Each tool comes with a certain set of requirements. For example, a
        Kubeflow installation will require you to have a Kubernetes cluster, and
        so would a Seldon Core deployment.


        🤔 Figuring out the defaults for infra parameters is not easy. Even if
        you have identified the backing infra that you need for a stack
        component, setting up reasonable defaults for parameters like instance
        size, CPU, memory, etc., needs a lot of experimentation to figure out.


        🚧 Many times, standard tool installations don't work out of the box.
        For example, to run a custom pipeline in Vertex AI, it is not enough to
        just run an imported pipeline. You might also need a custom service
        account that is configured to perform tasks like reading secrets from
        your secret store or talking to other GCP services that your pipeline
        might need.


        🔐 Some tools need an additional layer of installations to enable a more
        secure, production-grade setup. For example, a standard MLflow tracking
        server deployment comes without an authentication frontend which might
        expose all of your tracking data to the world if deployed as-is.
  - source_sentence: >-
      What is the expiration time for the GCP OAuth2 token in the ZenML
      configuration?
    sentences:
      - >-
        ━━━━━┛


        Configuration


        ┏━━━━━━━━━━━━┯━━━━━━━━━━━━┓┃ PROPERTY    VALUE      


        ┠────────────┼────────────┨


         project_id  zenml-core 


        ┠────────────┼────────────┨


         token       [HIDDEN]   


        ┗━━━━━━━━━━━━┷━━━━━━━━━━━━┛


        Note the temporary nature of the Service Connector. It will expire and
        become unusable in 1 hour:


        zenml service-connector list --name gcp-oauth2-token


        Example Command Output


        ┏━━━━━━━━┯━━━━━━━━━━━━━━━━━━┯━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┯━━━━━━━━┯━━━━━━━━━━━━━━━━━━━━━━━┯━━━━━━━━━━━━━━━┯━━━━━━━━┯━━━━━━━━━┯━━━━━━━━━━━━┯━━━━━━━━┓


         ACTIVE  NAME              ID                                   
        TYPE    RESOURCE TYPES         RESOURCE NAME  SHARED  OWNER   
        EXPIRES IN  LABELS 


        ┠────────┼──────────────────┼──────────────────────────────────────┼────────┼───────────────────────┼───────────────┼────────┼─────────┼────────────┼────────┨


                 gcp-oauth2-token  ec4d7d85-c71c-476b-aa76-95bf772c90da  🔵
        gcp  🔵 gcp-generic         <multiple>           default 
        59m35s             


                                                                       
                 📦 gcs-bucket                                        
                            


                                                                       
                 🌀 kubernetes-cluster                                
                            


                                                                       
                 🐳 docker-registry                                   
                            


        ┗━━━━━━━━┷━━━━━━━━━━━━━━━━━━┷━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┷━━━━━━━━┷━━━━━━━━━━━━━━━━━━━━━━━┷━━━━━━━━━━━━━━━┷━━━━━━━━┷━━━━━━━━━┷━━━━━━━━━━━━┷━━━━━━━━┛


        Auto-configuration


        The GCP Service Connector allows auto-discovering and fetching
        credentials and configuration set up by the GCP CLI on your local host.
      - >-
        Hugging Face


        Deploying models to Huggingface Inference Endpoints with Hugging Face
        :hugging_face:.


        Hugging Face Inference Endpoints provides a secure production solution
        to easily deploy any transformers, sentence-transformers, and diffusers
        models on a dedicated and autoscaling infrastructure managed by Hugging
        Face. An Inference Endpoint is built from a model from the Hub.


        This service provides dedicated and autoscaling infrastructure managed
        by Hugging Face, allowing you to deploy models without dealing with
        containers and GPUs.


        When to use it?


        You should use Hugging Face Model Deployer:


        if you want to deploy Transformers, Sentence-Transformers, or Diffusion
        models on dedicated and secure infrastructure.


        if you prefer a fully-managed production solution for inference without
        the need to handle containers and GPUs.


        if your goal is to turn your models into production-ready APIs with
        minimal infrastructure or MLOps involvement


        Cost-effectiveness is crucial, and you want to pay only for the raw
        compute resources you use.


        Enterprise security is a priority, and you need to deploy models into
        secure offline endpoints accessible only via a direct connection to your
        Virtual Private Cloud (VPCs).


        If you are looking for a more easy way to deploy your models locally,
        you can use the MLflow Model Deployer flavor.


        How to deploy it?


        The Hugging Face Model Deployer flavor is provided by the Hugging Face
        ZenML integration, so you need to install it on your local machine to be
        able to deploy your models. You can do this by running the following
        command:


        zenml integration install huggingface -y


        To register the Hugging Face model deployer with ZenML you need to run
        the following command:


        zenml model-deployer register <MODEL_DEPLOYER_NAME> --flavor=huggingface
        --token=<YOUR_HF_TOKEN> --namespace=<YOUR_HF_NAMESPACE>


        Here,


        token parameter is the Hugging Face authentication token. It can be
        managed through Hugging Face settings.
      - >-
        Can you list the steps to set up a Docker registry on a Kubernetes
        cluster?
model-index:
  - name: zenml/finetuned-snowflake-arctic-embed-m
    results:
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 384
          type: dim_384
        metrics:
          - type: cosine_accuracy@1
            value: 0.29518072289156627
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.5240963855421686
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.5843373493975904
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.6867469879518072
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.29518072289156627
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.17469879518072293
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.11686746987951804
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.0686746987951807
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.29518072289156627
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.5240963855421686
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.5843373493975904
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.6867469879518072
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.4908042072911187
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.42844234079173843
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.43576329240226386
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 256
          type: dim_256
        metrics:
          - type: cosine_accuracy@1
            value: 0.25903614457831325
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.5060240963855421
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.5783132530120482
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.6445783132530121
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.25903614457831325
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.1686746987951807
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.11566265060240961
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.0644578313253012
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.25903614457831325
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.5060240963855421
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.5783132530120482
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.6445783132530121
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.4548319777111225
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.39346194301013593
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.40343211538391555
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 128
          type: dim_128
        metrics:
          - type: cosine_accuracy@1
            value: 0.2710843373493976
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.46987951807228917
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.5662650602409639
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.6144578313253012
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.2710843373493976
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.1566265060240964
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.11325301204819276
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.061445783132530116
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.2710843373493976
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.46987951807228917
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.5662650602409639
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.6144578313253012
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.44433019669319024
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.3893574297188756
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.3989315479842741
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 64
          type: dim_64
        metrics:
          - type: cosine_accuracy@1
            value: 0.21686746987951808
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.42168674698795183
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.5180722891566265
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.5843373493975904
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.21686746987951808
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.14056224899598396
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.10361445783132528
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.05843373493975902
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.21686746987951808
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.42168674698795183
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.5180722891566265
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.5843373493975904
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.39639025659520544
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.3364529546758464
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.34658882510541217
            name: Cosine Map@100

zenml/finetuned-snowflake-arctic-embed-m

This is a sentence-transformers model finetuned from Snowflake/snowflake-arctic-embed-m. It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.

Model Details

Model Description

  • Model Type: Sentence Transformer
  • Base model: Snowflake/snowflake-arctic-embed-m
  • Maximum Sequence Length: 512 tokens
  • Output Dimensionality: 768 tokens
  • Similarity Function: Cosine Similarity
  • Language: en
  • License: apache-2.0

Model Sources

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel 
  (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
  (2): Normalize()
)

Usage

Direct Usage (Sentence Transformers)

First install the Sentence Transformers library:

pip install -U sentence-transformers

Then you can load this model and run inference.

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("zenml/finetuned-snowflake-arctic-embed-m")
# Run inference
sentences = [
    'What is the expiration time for the GCP OAuth2 token in the ZenML configuration?',
    '━━━━━┛\n\nConfiguration\n\n┏━━━━━━━━━━━━┯━━━━━━━━━━━━┓┃ PROPERTY   │ VALUE      ┃\n\n┠────────────┼────────────┨\n\n┃ project_id │ zenml-core ┃\n\n┠────────────┼────────────┨\n\n┃ token      │ [HIDDEN]   ┃\n\n┗━━━━━━━━━━━━┷━━━━━━━━━━━━┛\n\nNote the temporary nature of the Service Connector. It will expire and become unusable in 1 hour:\n\nzenml service-connector list --name gcp-oauth2-token\n\nExample Command Output\n\n┏━━━━━━━━┯━━━━━━━━━━━━━━━━━━┯━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┯━━━━━━━━┯━━━━━━━━━━━━━━━━━━━━━━━┯━━━━━━━━━━━━━━━┯━━━━━━━━┯━━━━━━━━━┯━━━━━━━━━━━━┯━━━━━━━━┓\n\n┃ ACTIVE │ NAME             │ ID                                   │ TYPE   │ RESOURCE TYPES        │ RESOURCE NAME │ SHARED │ OWNER   │ EXPIRES IN │ LABELS ┃\n\n┠────────┼──────────────────┼──────────────────────────────────────┼────────┼───────────────────────┼───────────────┼────────┼─────────┼────────────┼────────┨\n\n┃        │ gcp-oauth2-token │ ec4d7d85-c71c-476b-aa76-95bf772c90da │ 🔵 gcp │ 🔵 gcp-generic        │ <multiple>    │ ➖     │ default │ 59m35s     │        ┃\n\n┃        │                  │                                      │        │ 📦 gcs-bucket         │               │        │         │            │        ┃\n\n┃        │                  │                                      │        │ 🌀 kubernetes-cluster │               │        │         │            │        ┃\n\n┃        │                  │                                      │        │ 🐳 docker-registry    │               │        │         │            │        ┃\n\n┗━━━━━━━━┷━━━━━━━━━━━━━━━━━━┷━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┷━━━━━━━━┷━━━━━━━━━━━━━━━━━━━━━━━┷━━━━━━━━━━━━━━━┷━━━━━━━━┷━━━━━━━━━┷━━━━━━━━━━━━┷━━━━━━━━┛\n\nAuto-configuration\n\nThe GCP Service Connector allows auto-discovering and fetching credentials and configuration set up by the GCP CLI on your local host.',
    'Can you list the steps to set up a Docker registry on a Kubernetes cluster?',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]

Evaluation

Metrics

Information Retrieval

Metric Value
cosine_accuracy@1 0.2952
cosine_accuracy@3 0.5241
cosine_accuracy@5 0.5843
cosine_accuracy@10 0.6867
cosine_precision@1 0.2952
cosine_precision@3 0.1747
cosine_precision@5 0.1169
cosine_precision@10 0.0687
cosine_recall@1 0.2952
cosine_recall@3 0.5241
cosine_recall@5 0.5843
cosine_recall@10 0.6867
cosine_ndcg@10 0.4908
cosine_mrr@10 0.4284
cosine_map@100 0.4358

Information Retrieval

Metric Value
cosine_accuracy@1 0.259
cosine_accuracy@3 0.506
cosine_accuracy@5 0.5783
cosine_accuracy@10 0.6446
cosine_precision@1 0.259
cosine_precision@3 0.1687
cosine_precision@5 0.1157
cosine_precision@10 0.0645
cosine_recall@1 0.259
cosine_recall@3 0.506
cosine_recall@5 0.5783
cosine_recall@10 0.6446
cosine_ndcg@10 0.4548
cosine_mrr@10 0.3935
cosine_map@100 0.4034

Information Retrieval

Metric Value
cosine_accuracy@1 0.2711
cosine_accuracy@3 0.4699
cosine_accuracy@5 0.5663
cosine_accuracy@10 0.6145
cosine_precision@1 0.2711
cosine_precision@3 0.1566
cosine_precision@5 0.1133
cosine_precision@10 0.0614
cosine_recall@1 0.2711
cosine_recall@3 0.4699
cosine_recall@5 0.5663
cosine_recall@10 0.6145
cosine_ndcg@10 0.4443
cosine_mrr@10 0.3894
cosine_map@100 0.3989

Information Retrieval

Metric Value
cosine_accuracy@1 0.2169
cosine_accuracy@3 0.4217
cosine_accuracy@5 0.5181
cosine_accuracy@10 0.5843
cosine_precision@1 0.2169
cosine_precision@3 0.1406
cosine_precision@5 0.1036
cosine_precision@10 0.0584
cosine_recall@1 0.2169
cosine_recall@3 0.4217
cosine_recall@5 0.5181
cosine_recall@10 0.5843
cosine_ndcg@10 0.3964
cosine_mrr@10 0.3365
cosine_map@100 0.3466

Training Details

Training Dataset

Unnamed Dataset

  • Size: 1,490 training samples
  • Columns: positive, anchor, and negative
  • Approximate statistics based on the first 1000 samples:
    positive anchor negative
    type string string string
    details
    • min: 9 tokens
    • mean: 21.02 tokens
    • max: 64 tokens
    • min: 23 tokens
    • mean: 375.16 tokens
    • max: 512 tokens
    • min: 10 tokens
    • mean: 17.51 tokens
    • max: 31 tokens
  • Samples:
    positive anchor negative
    What details can you provide about the mlflow_training_pipeline runs listed in the ZenML documentation? mlflow_training_pipeline', ┃┃ │ │ │ 'zenml_pipeline_run_uuid': 'a5d4faae-ef70-48f2-9893-6e65d5e51e98', 'zenml_workspace': '10e060b3-2f7e-463d-9ec8-3a211ef4e1f6', 'epochs': '5', 'optimizer': 'Adam', 'lr': '0.005'} ┃

    ┠────────────────────────┼───────────────┼─────────────────────────────────────────┼──────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────┨

    ┃ tensorflow-mnist-model │ 2 │ Run #2 of the mlflow_training_pipeline. │ {'zenml_version': '0.34.0', 'zenml_run_name': 'mlflow_training_pipeline-2023_03_01-08_09_08_467212', 'zenml_pipeline_name': 'mlflow_training_pipeline', ┃

    ┃ │ │ │ 'zenml_pipeline_run_uuid': '11858dcf-3e47-4b1a-82c5-6fa25ba4e037', 'zenml_workspace': '10e060b3-2f7e-463d-9ec8-3a211ef4e1f6', 'epochs': '5', 'optimizer': 'Adam', 'lr': '0.003'} ┃

    ┠────────────────────────┼───────────────┼─────────────────────────────────────────┼──────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────┨

    ┃ tensorflow-mnist-model │ 1 │ Run #1 of the mlflow_training_pipeline. │ {'zenml_version': '0.34.0', 'zenml_run_name': 'mlflow_training_pipeline-2023_03_01-08_08_52_398499', 'zenml_pipeline_name': 'mlflow_training_pipeline', ┃

    ┃ │ │ │ 'zenml_pipeline_run_uuid': '29fb22c1-6e0b-4431-9e04-226226506d16', 'zenml_workspace': '10e060b3-2f7e-463d-9ec8-3a211ef4e1f6', 'epochs': '5', 'optimizer': 'Adam', 'lr': '0.001'} ┃
    Can you explain how to configure the TensorFlow settings for a different project?
    How do you register a GCP Service Connector that uses account impersonation to access the zenml-bucket-sl GCS bucket? esource-id zenml-bucket-sl

    Example Command OutputError: Service connector 'gcp-empty-sa' verification failed: connector authorization failure: failed to fetch GCS bucket

    zenml-bucket-sl: 403 GET https://storage.googleapis.com/storage/v1/b/zenml-bucket-sl?projection=noAcl&prettyPrint=false:

    [email protected] does not have storage.buckets.get access to the Google Cloud Storage bucket.

    Permission 'storage.buckets.get' denied on resource (or it may not exist).

    Next, we'll register a GCP Service Connector that actually uses account impersonation to access the zenml-bucket-sl GCS bucket and verify that it can actually access the bucket:

    zenml service-connector register gcp-impersonate-sa --type gcp --auth-method impersonation --service_account_json=@[email protected] --project_id=zenml-core --target_principal=[email protected] --resource-type gcs-bucket --resource-id gs://zenml-bucket-sl

    Example Command Output

    Expanding argument value service_account_json to contents of file /home/stefan/aspyre/src/zenml/[email protected].

    Successfully registered service connector gcp-impersonate-sa with access to the following resources:

    ┏━━━━━━━━━━━━━━━┯━━━━━━━━━━━━━━━━━━━━━━┓

    ┃ RESOURCE TYPE │ RESOURCE NAMES ┃

    ┠───────────────┼──────────────────────┨

    ┃ 📦 gcs-bucket │ gs://zenml-bucket-sl ┃

    ┗━━━━━━━━━━━━━━━┷━━━━━━━━━━━━━━━━━━━━━━┛

    External Account (GCP Workload Identity)

    Use GCP workload identity federation to authenticate to GCP services using AWS IAM credentials, Azure Active Directory credentials or generic OIDC tokens.
    What is the process for setting up a ZenML pipeline using AWS IAM credentials?
    Can you explain how data validation helps in detecting data drift and model drift in ZenML pipelines? of your models at different stages of development.if you have pipelines that regularly ingest new data, you should use data validation to run regular data integrity checks to signal problems before they are propagated downstream.

    in continuous training pipelines, you should use data validation techniques to compare new training data against a data reference and to compare the performance of newly trained models against previous ones.

    when you have pipelines that automate batch inference or if you regularly collect data used as input in online inference, you should use data validation to run data drift analyses and detect training-serving skew, data drift and model drift.

    Data Validator Flavors

    Data Validator are optional stack components provided by integrations. The following table lists the currently available Data Validators and summarizes their features and the data types and model types that they can be used with in ZenML pipelines:

    Data Validator Validation Features Data Types Model Types Notes Flavor/Integration Deepchecks data quality
    data drift
    model drift
    model performance tabular: pandas.DataFrame CV: torch.utils.data.dataloader.DataLoader tabular: sklearn.base.ClassifierMixin CV: torch.nn.Module Add Deepchecks data and model validation tests to your pipelines deepchecks Evidently data quality
    data drift
    model drift
    model performance tabular: pandas.DataFrame N/A Use Evidently to generate a variety of data quality and data/model drift reports and visualizations evidently Great Expectations data profiling
    data quality tabular: pandas.DataFrame N/A Perform data testing, documentation and profiling with Great Expectations great_expectations Whylogs/WhyLabs data drift tabular: pandas.DataFrame N/A Generate data profiles with whylogs and upload them to WhyLabs whylogs

    If you would like to see the available flavors of Data Validator, you can use the command:

    zenml data-validator flavor list

    How to use it
    What are the best practices for deploying web applications using Docker and Kubernetes?
  • Loss: MatryoshkaLoss with these parameters:
    {
        "loss": "TripletLoss",
        "matryoshka_dims": [
            384,
            256,
            128,
            64
        ],
        "matryoshka_weights": [
            1,
            1,
            1,
            1
        ],
        "n_dims_per_step": -1
    }
    

Training Hyperparameters

Non-Default Hyperparameters

  • eval_strategy: epoch
  • per_device_train_batch_size: 32
  • per_device_eval_batch_size: 16
  • gradient_accumulation_steps: 16
  • learning_rate: 2e-05
  • num_train_epochs: 4
  • lr_scheduler_type: cosine
  • warmup_ratio: 0.1
  • bf16: True
  • tf32: True
  • load_best_model_at_end: True
  • optim: adamw_torch_fused
  • batch_sampler: no_duplicates

All Hyperparameters

Click to expand
  • overwrite_output_dir: False
  • do_predict: False
  • eval_strategy: epoch
  • prediction_loss_only: True
  • per_device_train_batch_size: 32
  • per_device_eval_batch_size: 16
  • per_gpu_train_batch_size: None
  • per_gpu_eval_batch_size: None
  • gradient_accumulation_steps: 16
  • eval_accumulation_steps: None
  • learning_rate: 2e-05
  • weight_decay: 0.0
  • adam_beta1: 0.9
  • adam_beta2: 0.999
  • adam_epsilon: 1e-08
  • max_grad_norm: 1.0
  • num_train_epochs: 4
  • max_steps: -1
  • lr_scheduler_type: cosine
  • lr_scheduler_kwargs: {}
  • warmup_ratio: 0.1
  • warmup_steps: 0
  • log_level: passive
  • log_level_replica: warning
  • log_on_each_node: True
  • logging_nan_inf_filter: True
  • save_safetensors: True
  • save_on_each_node: False
  • save_only_model: False
  • restore_callback_states_from_checkpoint: False
  • no_cuda: False
  • use_cpu: False
  • use_mps_device: False
  • seed: 42
  • data_seed: None
  • jit_mode_eval: False
  • use_ipex: False
  • bf16: True
  • fp16: False
  • fp16_opt_level: O1
  • half_precision_backend: auto
  • bf16_full_eval: False
  • fp16_full_eval: False
  • tf32: True
  • local_rank: 0
  • ddp_backend: None
  • tpu_num_cores: None
  • tpu_metrics_debug: False
  • debug: []
  • dataloader_drop_last: False
  • dataloader_num_workers: 0
  • dataloader_prefetch_factor: None
  • past_index: -1
  • disable_tqdm: True
  • remove_unused_columns: True
  • label_names: None
  • load_best_model_at_end: True
  • ignore_data_skip: False
  • fsdp: []
  • fsdp_min_num_params: 0
  • fsdp_config: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
  • fsdp_transformer_layer_cls_to_wrap: None
  • accelerator_config: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
  • deepspeed: None
  • label_smoothing_factor: 0.0
  • optim: adamw_torch_fused
  • optim_args: None
  • adafactor: False
  • group_by_length: False
  • length_column_name: length
  • ddp_find_unused_parameters: None
  • ddp_bucket_cap_mb: None
  • ddp_broadcast_buffers: False
  • dataloader_pin_memory: True
  • dataloader_persistent_workers: False
  • skip_memory_metrics: True
  • use_legacy_prediction_loop: False
  • push_to_hub: False
  • resume_from_checkpoint: None
  • hub_model_id: None
  • hub_strategy: every_save
  • hub_private_repo: False
  • hub_always_push: False
  • gradient_checkpointing: False
  • gradient_checkpointing_kwargs: None
  • include_inputs_for_metrics: False
  • eval_do_concat_batches: True
  • fp16_backend: auto
  • push_to_hub_model_id: None
  • push_to_hub_organization: None
  • mp_parameters:
  • auto_find_batch_size: False
  • full_determinism: False
  • torchdynamo: None
  • ray_scope: last
  • ddp_timeout: 1800
  • torch_compile: False
  • torch_compile_backend: None
  • torch_compile_mode: None
  • dispatch_batches: None
  • split_batches: None
  • include_tokens_per_second: False
  • include_num_input_tokens_seen: False
  • neftune_noise_alpha: None
  • optim_target_modules: None
  • batch_eval_metrics: False
  • batch_sampler: no_duplicates
  • multi_dataset_batch_sampler: proportional

Training Logs

Epoch Step dim_128_cosine_map@100 dim_256_cosine_map@100 dim_384_cosine_map@100 dim_64_cosine_map@100
0.6667 1 0.3884 0.4332 0.4464 0.3140
2.0 3 0.4064 0.4195 0.4431 0.3553
2.6667 4 0.3989 0.4034 0.4358 0.3466
  • The bold row denotes the saved checkpoint.

Framework Versions

  • Python: 3.10.14
  • Sentence Transformers: 3.0.1
  • Transformers: 4.41.2
  • PyTorch: 2.3.1+cu121
  • Accelerate: 0.31.0
  • Datasets: 2.19.1
  • Tokenizers: 0.19.1

Citation

BibTeX

Sentence Transformers

@inproceedings{reimers-2019-sentence-bert,
    title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
    author = "Reimers, Nils and Gurevych, Iryna",
    booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
    month = "11",
    year = "2019",
    publisher = "Association for Computational Linguistics",
    url = "https://arxiv.org/abs/1908.10084",
}

MatryoshkaLoss

@misc{kusupati2024matryoshka,
    title={Matryoshka Representation Learning}, 
    author={Aditya Kusupati and Gantavya Bhatt and Aniket Rege and Matthew Wallingford and Aditya Sinha and Vivek Ramanujan and William Howard-Snyder and Kaifeng Chen and Sham Kakade and Prateek Jain and Ali Farhadi},
    year={2024},
    eprint={2205.13147},
    archivePrefix={arXiv},
    primaryClass={cs.LG}
}

TripletLoss

@misc{hermans2017defense,
    title={In Defense of the Triplet Loss for Person Re-Identification}, 
    author={Alexander Hermans and Lucas Beyer and Bastian Leibe},
    year={2017},
    eprint={1703.07737},
    archivePrefix={arXiv},
    primaryClass={cs.CV}
}