Users of the MLflow Open Source Project who are hosting the MLflow Model Registry using the mlflow server or mlflow ui commands using an MLflow version older than MLflow 2.2.1 may be vulnerable to a remote file existence check exploit if they are not limiting who can query their server (for example, by using a cloud VPC, an IP allowlist for inbound requests, or authentication / authorization middleware).
This issue only affects users and integrations that run the mlflow server and mlflow ui commands. Integrations that do not make use of mlflow server or mlflow ui are unaffected; for example, the Databricks Managed MLflow product and MLflow on Azure Machine Learning do not make use of these commands and are not impacted by these vulnerabilities in any way.
The vulnerability detailed in https://nvd.nist.gov/vuln/detail/CVE-2023-1176 enables an actor to check the existence of arbitrary files unrelated to MLflow from the host server, including any files stored in remote locations to which the host server has access.
Patches
This vulnerability has been patched in MLflow 2.2.1, which was released to PyPI on March 2nd, 2023. If you are using mlflow server or mlflow ui with the MLflow Model Registry, we recommend upgrading to MLflow 2.2.1 as soon as possible.
Workarounds
If you are using the MLflow open source mlflow server or mlflow ui commands, we strongly recommend limiting who can access your MLflow Model Registry and MLflow Tracking servers using a cloud VPC, an IP allowlist for inbound requests, authentication / authorization middleware, or another access restriction mechanism of your choosing.
If you are using the MLflow open source mlflow server or mlflow ui commands, we also strongly recommend limiting the remote files to which your MLflow Model Registry and MLflow Tracking servers have access. For example, if your MLflow Model Registry or MLflow Tracking server uses cloud-hosted blob storage for MLflow artifacts, make sure to restrict the scope of your server's cloud credentials such that it can only access files and directories related to MLflow.
Users of the MLflow Open Source Project who are hosting the MLflow Model Registry using the mlflow server or mlflow ui commands using an MLflow version older than MLflow 2.2.1 may be vulnerable to a remote file access exploit if they are not limiting who can query their server (for example, by using a cloud VPC, an IP allowlist for inbound requests, or authentication / authorization middleware).
This issue only affects users and integrations that run the mlflow server and mlflow ui commands. Integrations that do not make use of mlflow server or mlflow ui are unaffected; for example, the Databricks Managed MLflow product and MLflow on Azure Machine Learning do not make use of these commands and are not impacted by these vulnerabilities in any way.
The vulnerability detailed in https://nvd.nist.gov/vuln/detail/CVE-2023-1177 enables an actor to download arbitrary files unrelated to MLflow from the host server, including any files stored in remote locations to which the host server has access.
Patches
This vulnerability has been patched in MLflow 2.2.1, which was released to PyPI on March 2nd, 2023. If you are using mlflow server or mlflow ui with the MLflow Model Registry, we recommend upgrading to MLflow 2.2.1 as soon as possible.
Workarounds
If you are using the MLflow open source mlflow server or mlflow ui commands, we strongly recommend limiting who can access your MLflow Model Registry and MLflow Tracking servers using a cloud VPC, an IP allowlist for inbound requests, authentication / authorization middleware, or another access restriction mechanism of your choosing.
If you are using the MLflow open source mlflow server or mlflow ui commands, we also strongly recommend limiting the remote files to which your MLflow Model Registry and MLflow Tracking servers have access. For example, if your MLflow Model Registry or MLflow Tracking server uses cloud-hosted blob storage for MLflow artifacts, make sure to restrict the scope of your server's cloud credentials such that it can only access files and directories related to MLflow.
Users of the MLflow Open Source Project who are hosting the MLflow Model Registry using the mlflow server or mlflow ui commands using an MLflow version older than MLflow 2.3.1 may be vulnerable to a remote file access exploit if they are not limiting who can query their server (for example, by using a cloud VPC, an IP allowlist for inbound requests, or authentication / authorization middleware).
This issue only affects users and integrations that run the mlflow server and mlflow ui commands. Integrations that do not make use of mlflow server or mlflow ui are unaffected; for example, the Databricks Managed MLflow product and MLflow on Azure Machine Learning do not make use of these commands and are not impacted by these vulnerabilities in any way.
This vulnerability has been patched in MLflow 2.3.1, which was released to PyPI on April 27th, 2023. If you are using mlflow server or mlflow ui with the MLflow Model Registry, we recommend upgrading to MLflow 2.3.1 as soon as possible.
Workarounds
If you are using the MLflow open source mlflow server or mlflow ui commands, we strongly recommend limiting who can access your MLflow Model Registry and MLflow Tracking servers using a cloud VPC, an IP allowlist for inbound requests, authentication / authorization middleware, or another access restriction mechanism of your choosing.
If you are using the MLflow open source mlflow server or mlflow ui commands, we also strongly recommend limiting the remote files to which your MLflow Model Registry and MLflow Tracking servers have access. For example, if your MLflow Model Registry or MLflow Tracking server uses cloud-hosted blob storage for MLflow artifacts, make sure to restrict the scope of your server's cloud credentials such that it can only access files and directories related to MLflow.
A directory traversal vulnerability in the /get-artifact API method of the mlflow platform prior to v2.0.0 allows attackers to read arbitrary files on the server via the path parameter.
Absolute Path Traversal in GitHub repository mlflow/mlflow prior to 2.5.0.
Release Notes
mlflow/mlflow (mlflow)
### [`v2.5.0`](https://togithub.com/mlflow/mlflow/blob/HEAD/CHANGELOG.md#250-2023-07-17)
[Compare Source](https://togithub.com/mlflow/mlflow/compare/v2.4.2...v2.5.0)
MLflow 2.5.0 includes several major features and improvements:
- \[MLflow AI Gateway] We are excited to announce the release of MLflow AI Gateway, a powerful tool designed to streamline the usage and management of various large language model (LLM) providers, such as OpenAI and Anthropic, within an organization. It offers a standardized interface that simplifies the interaction with these services and delivers centralized, secure management of credentials. To get started with MLflow AI Gateway, check out the docs at https://mlflow.org/docs/latest/gateway/index.html. ([#8694](https://togithub.com/mlflow/mlflow/issues/8694), [@harupy](https://togithub.com/harupy), [@BenWilson2](https://togithub.com/BenWilson2), [@dbczumar](https://togithub.com/dbczumar))
- \[Auth]: We are excited to announce the release of authentication and authorization support for MLflow Tracking and the MLflow Model Registry, providing integrated access control capabilities to both services. To get started, check out the docs at https://mlflow.org/docs/latest/auth/index.html. ([#9000](https://togithub.com/mlflow/mlflow/issues/9000), [#8975](https://togithub.com/mlflow/mlflow/issues/8975), [#8626](https://togithub.com/mlflow/mlflow/issues/8626), [#8837](https://togithub.com/mlflow/mlflow/issues/8837), [#8841](https://togithub.com/mlflow/mlflow/issues/8841), [@gabrielfu](https://togithub.com/gabrielfu), [@harupy](https://togithub.com/harupy))
Features:
- \[Models] Add Support to the LangChain flavor for chains that contain unserializable components ([#8736](https://togithub.com/mlflow/mlflow/issues/8736), [@liangz1](https://togithub.com/liangz1))
- \[Scoring] Infer spark udf return type from model output schema ([#8934](https://togithub.com/mlflow/mlflow/issues/8934), [@WeichenXu123](https://togithub.com/WeichenXu123))
- \[Models] Add support for automated signature inference ([#8860](https://togithub.com/mlflow/mlflow/issues/8860), [#8782](https://togithub.com/mlflow/mlflow/issues/8782) [#8795](https://togithub.com/mlflow/mlflow/issues/8795), [#8725](https://togithub.com/mlflow/mlflow/issues/8725), [@jerrylian-db](https://togithub.com/jerrylian-db))
Bug fixes:
- \[Security] Improve robustness to LFI attacks on Windows by enhancing path validation ([#8999](https://togithub.com/mlflow/mlflow/issues/8999), [@serena-ruan](https://togithub.com/serena-ruan))
- If you are using `mlflow server` or `mlflow ui` on Windows, we recommend upgrading to MLflow 2.5.0 as soon as possible.
- \[Scoring] Support nullable array type values as spark_udf return values ([#9014](https://togithub.com/mlflow/mlflow/issues/9014), [@WeichenXu123](https://togithub.com/WeichenXu123))
- \[Models] Revert cache deletion of system modules when adding custom model code to the system path ([#8722](https://togithub.com/mlflow/mlflow/issues/8722), [@trungn1](https://togithub.com/trungn1))
- \[Models] add micro version to mlflow version pinning ([#8687](https://togithub.com/mlflow/mlflow/issues/8687), [@C-K-Loan](https://togithub.com/C-K-Loan))
- \[Artifacts] Prevent manually deleted artifacts from causing artifact garbage collection to fail ([#8498](https://togithub.com/mlflow/mlflow/issues/8498), [@PenHsuanWang](https://togithub.com/PenHsuanWang))
Documentation updates:
- \[Docs] Update .push_model_to_sagemaker docs ([#8851](https://togithub.com/mlflow/mlflow/issues/8851), [@pdifranc](https://togithub.com/pdifranc))
- \[Docs] Fix invalid link for Azure ML documentation ([#8800](https://togithub.com/mlflow/mlflow/issues/8800), [@dunnkers](https://togithub.com/dunnkers))
- \[Artifacts / Docs / Models / Projects] Adds information on the OCI MLflow plugins for seamless integration with Oralce Cloud Infrastructure services. ([#8707](https://togithub.com/mlflow/mlflow/issues/8707), [@mrDzurb](https://togithub.com/mrDzurb))
Deprecation:
- \[Models] Deprecate the `gluon` model flavor. The `mlflow.gluon` module will be removed in a future release. ([#8968](https://togithub.com/mlflow/mlflow/issues/8968), [@harupy](https://togithub.com/harupy))
Small bug fixes and documentation updates:
[#9069](https://togithub.com/mlflow/mlflow/issues/9069), [#9056](https://togithub.com/mlflow/mlflow/issues/9056), [#9055](https://togithub.com/mlflow/mlflow/issues/9055), [#9054](https://togithub.com/mlflow/mlflow/issues/9054), [#9048](https://togithub.com/mlflow/mlflow/issues/9048), [#9043](https://togithub.com/mlflow/mlflow/issues/9043), [#9035](https://togithub.com/mlflow/mlflow/issues/9035), [#9034](https://togithub.com/mlflow/mlflow/issues/9034), [#9037](https://togithub.com/mlflow/mlflow/issues/9037), [#9038](https://togithub.com/mlflow/mlflow/issues/9038), [#8993](https://togithub.com/mlflow/mlflow/issues/8993), [#8966](https://togithub.com/mlflow/mlflow/issues/8966), [#8985](https://togithub.com/mlflow/mlflow/issues/8985), [@BenWilson2](https://togithub.com/BenWilson2); [#9039](https://togithub.com/mlflow/mlflow/issues/9039), [#9036](https://togithub.com/mlflow/mlflow/issues/9036), [#8902](https://togithub.com/mlflow/mlflow/issues/8902), [#8924](https://togithub.com/mlflow/mlflow/issues/8924), [#8866](https://togithub.com/mlflow/mlflow/issues/8866), [#8861](https://togithub.com/mlflow/mlflow/issues/8861), [#8810](https://togithub.com/mlflow/mlflow/issues/8810), [#8761](https://togithub.com/mlflow/mlflow/issues/8761), [#8544](https://togithub.com/mlflow/mlflow/issues/8544), [@jerrylian-db](https://togithub.com/jerrylian-db); [#8903](https://togithub.com/mlflow/mlflow/issues/8903), [@smurching](https://togithub.com/smurching); [#9080](https://togithub.com/mlflow/mlflow/issues/9080), [#9079](https://togithub.com/mlflow/mlflow/issues/9079), [#9078](https://togithub.com/mlflow/mlflow/issues/9078), [#9076](https://togithub.com/mlflow/mlflow/issues/9076), [#9075](https://togithub.com/mlflow/mlflow/issues/9075), [#9074](https://togithub.com/mlflow/mlflow/issues/9074), [#9071](https://togithub.com/mlflow/mlflow/issues/9071), [#9063](https://togithub.com/mlflow/mlflow/issues/9063), [#9062](https://togithub.com/mlflow/mlflow/issues/9062), [#9032](https://togithub.com/mlflow/mlflow/issues/9032), [#9031](https://togithub.com/mlflow/mlflow/issues/9031), [#9027](https://togithub.com/mlflow/mlflow/issues/9027), [#9023](https://togithub.com/mlflow/mlflow/issues/9023), [#9022](https://togithub.com/mlflow/mlflow/issues/9022), [#9020](https://togithub.com/mlflow/mlflow/issues/9020), [#9005](https://togithub.com/mlflow/mlflow/issues/9005), [#8994](https://togithub.com/mlflow/mlflow/issues/8994), [#8979](https://togithub.com/mlflow/mlflow/issues/8979), [#8983](https://togithub.com/mlflow/mlflow/issues/8983), [#8984](https://togithub.com/mlflow/mlflow/issues/8984), [#8982](https://togithub.com/mlflow/mlflow/issues/8982), [#8970](https://togithub.com/mlflow/mlflow/issues/8970), [#8962](https://togithub.com/mlflow/mlflow/issues/8962), [#8969](https://togithub.com/mlflow/mlflow/issues/8969), [#8968](https://togithub.com/mlflow/mlflow/issues/8968), [#8959](https://togithub.com/mlflow/mlflow/issues/8959), [#8960](https://togithub.com/mlflow/mlflow/issues/8960), [#8958](https://togithub.com/mlflow/mlflow/issues/8958), [#8956](https://togithub.com/mlflow/mlflow/issues/8956), [#8955](https://togithub.com/mlflow/mlflow/issues/8955), [#8954](https://togithub.com/mlflow/mlflow/issues/8954), [#8949](https://togithub.com/mlflow/mlflow/issues/8949), [#8950](https://togithub.com/mlflow/mlflow/issues/8950), [#8952](https://togithub.com/mlflow/mlflow/issues/8952), [#8948](https://togithub.com/mlflow/mlflow/issues/8948), [#8946](https://togithub.com/mlflow/mlflow/issues/8946), [#8947](https://togithub.com/mlflow/mlflow/issues/8947), [#8943](https://togithub.com/mlflow/mlflow/issues/8943), [#8944](https://togithub.com/mlflow/mlflow/issues/8944), [#8916](https://togithub.com/mlflow/mlflow/issues/8916), [#8917](https://togithub.com/mlflow/mlflow/issues/8917), [#8933](https://togithub.com/mlflow/mlflow/issues/8933), [#8929](https://togithub.com/mlflow/mlflow/issues/8929), [#8932](https://togithub.com/mlflow/mlflow/issues/8932), [#8927](https://togithub.com/mlflow/mlflow/issues/8927), [#8930](https://togithub.com/mlflow/mlflow/issues/8930), [#8925](https://togithub.com/mlflow/mlflow/issues/8925), [#8921](https://togithub.com/mlflow/mlflow/issues/8921), [#8873](https://togithub.com/mlflow/mlflow/issues/8873), [#8915](https://togithub.com/mlflow/mlflow/issues/8915), [#8909](https://togithub.com/mlflow/mlflow/issues/8909), [#8908](https://togithub.com/mlflow/mlflow/issues/8908), [#8911](https://togithub.com/mlflow/mlflow/issues/8911), [#8910](https://togithub.com/mlflow/mlflow/issues/8910), [#8907](https://togithub.com/mlflow/mlflow/issues/8907), [#8906](https://togithub.com/mlflow/mlflow/issues/8906), [#8898](https://togithub.com/mlflow/mlflow/issues/8898), [#8893](https://togithub.com/mlflow/mlflow/issues/8893), [#8889](https://togithub.com/mlflow/mlflow/issues/8889), [#8892](https://togithub.com/mlflow/mlflow/issues/8892), [#8891](https://togithub.com/mlflow/mlflow/issues/8891), [#8887](https://togithub.com/mlflow/mlflow/issues/8887), [#8875](https://togithub.com/mlflow/mlflow/issues/8875), [#8876](https://togithub.com/mlflow/mlflow/issues/8876), [#8882](https://togithub.com/mlflow/mlflow/issues/8882), [#8874](https://togithub.com/mlflow/mlflow/issues/8874), [#8868](https://togithub.com/mlflow/mlflow/issues/8868), [#8872](https://togithub.com/mlflow/mlflow/issues/8872), [#8869](https://togithub.com/mlflow/mlflow/issues/8869), [#8828](https://togithub.com/mlflow/mlflow/issues/8828), [#8852](https://togithub.com/mlflow/mlflow/issues/8852), [#8857](https://togithub.com/mlflow/mlflow/issues/8857), [#8853](https://togithub.com/mlflow/mlflow/issues/8853), [#8854](https://togithub.com/mlflow/mlflow/issues/8854), [#8848](https://togithub.com/mlflow/mlflow/issues/8848), [#8850](https://togithub.com/mlflow/mlflow/issues/8850), [#8840](https://togithub.com/mlflow/mlflow/issues/8840), [#8835](https://togithub.com/mlflow/mlflow/issues/8835), [#8832](https://togithub.com/mlflow/mlflow/issues/8832), [#8831](https://togithub.com/mlflow/mlflow/issues/8831), [#8830](https://togithub.com/mlflow/mlflow/issues/8830), [#8829](https://togithub.com/mlflow/mlflow/issues/8829), [#8839](https://togithub.com/mlflow/mlflow/issues/8839), [#8833](https://togithub.com/mlflow/mlflow/issues/8833), [#8838](https://togithub.com/mlflow/mlflow/issues/8838), [#8819](https://togithub.com/mlflow/mlflow/issues/8819), [#8814](https://togithub.com/mlflow/mlflow/issues/8814), [#8825](https://togithub.com/mlflow/mlflow/issues/8825), [#8818](https://togithub.com/mlflow/mlflow/issues/8818), [#8787](https://togithub.com/mlflow/mlflow/issues/8787), [#8775](https://togithub.com/mlflow/mlflow/issues/8775), [#8749](https://togithub.com/mlflow/mlflow/issues/8749), [#8766](https://togithub.com/mlflow/mlflow/issues/8766), [#8756](https://togithub.com/mlflow/mlflow/issues/8756), [#8753](https://togithub.com/mlflow/mlflow/issues/8753), [#8751](https://togithub.com/mlflow/mlflow/issues/8751), [#8748](https://togithub.com/mlflow/mlflow/issues/8748), [#8744](https://togithub.com/mlflow/mlflow/issues/8744), [#8731](https://togithub.com/mlflow/mlflow/issues/8731), [#8717](https://togithub.com/mlflow/mlflow/issues/8717), [#8730](https://togithub.com/mlflow/mlflow/issues/8730), [#8691](https://togithub.com/mlflow/mlflow/issues/8691), [#8720](https://togithub.com/mlflow/mlflow/issues/8720), [#8723](https://togithub.com/mlflow/mlflow/issues/8723), [#8719](https://togithub.com/mlflow/mlflow/issues/8719), [#8688](https://togithub.com/mlflow/mlflow/issues/8688), [#8721](https://togithub.com/mlflow/mlflow/issues/8721), [#8715](https://togithub.com/mlflow/mlflow/issues/8715), [#8716](https://togithub.com/mlflow/mlflow/issues/8716), [#8718](https://togithub.com/mlflow/mlflow/issues/8718), [#8696](https://togithub.com/mlflow/mlflow/issues/8696), [#8698](https://togithub.com/mlflow/mlflow/issues/8698), [#8692](https://togithub.com/mlflow/mlflow/issues/8692), [#8693](https://togithub.com/mlflow/mlflow/issues/8693), [#8690](https://togithub.com/mlflow/mlflow/issues/8690), [@harupy](https://togithub.com/harupy); [#9030](https://togithub.com/mlflow/mlflow/issues/9030), [@AlimurtuzaCodes](https://togithub.com/AlimurtuzaCodes); [#9029](https://togithub.com/mlflow/mlflow/issues/9029), [#9025](https://togithub.com/mlflow/mlflow/issues/9025), [#9021](https://togithub.com/mlflow/mlflow/issues/9021), [#9013](https://togithub.com/mlflow/mlflow/issues/9013), [@viktoriussuwandi](https://togithub.com/viktoriussuwandi); [#9010](https://togithub.com/mlflow/mlflow/issues/9010), [@Bncer](https://togithub.com/Bncer); [#9011](https://togithub.com/mlflow/mlflow/issues/9011), [@Pecunia201](https://togithub.com/Pecunia201); [#9007](https://togithub.com/mlflow/mlflow/issues/9007), [#9003](https://togithub.com/mlflow/mlflow/issues/9003), [@EdAbati](https://togithub.com/EdAbati); [#9002](https://togithub.com/mlflow/mlflow/issues/9002), [@prithvikannan](https://togithub.com/prithvikannan); [#8991](https://togithub.com/mlflow/mlflow/issues/8991), [#8867](https://togithub.com/mlflow/mlflow/issues/8867), [@AveshCSingh](https://togithub.com/AveshCSingh); [#8951](https://togithub.com/mlflow/mlflow/issues/8951), [#8896](https://togithub.com/mlflow/mlflow/issues/8896), [#8888](https://togithub.com/mlflow/mlflow/issues/8888), [#8849](https://togithub.com/mlflow/mlflow/issues/8849), [@gabrielfu](https://togithub.com/gabrielfu); [#8913](https://togithub.com/mlflow/mlflow/issues/8913), [#8885](https://togithub.com/mlflow/mlflow/issues/8885), [#8871](https://togithub.com/mlflow/mlflow/issues/8871), [#8870](https://togithub.com/mlflow/mlflow/issues/8870), [#8788](https://togithub.com/mlflow/mlflow/issues/8788), [#8772](https://togithub.com/mlflow/mlflow/issues/8772), [#8771](https://togithub.com/mlflow/mlflow/issues/8771), [@serena-ruan](https://togithub.com/serena-ruan); [#8879](https://togithub.com/mlflow/mlflow/issues/8879), [@maciejskorski](https://togithub.com/maciejskorski); [#7752](https://togithub.com/mlflow/mlflow/issues/7752), [@arunkumarkota](https://togithub.com/arunkumarkota); [#9083](https://togithub.com/mlflow/mlflow/issues/9083), [#9081](https://togithub.com/mlflow/mlflow/issues/9081), [#8765](https://togithub.com/mlflow/mlflow/issues/8765), [#8742](https://togithub.com/mlflow/mlflow/issues/8742), [#8685](https://togithub.com/mlflow/mlflow/issues/8685), [#8682](https://togithub.com/mlflow/mlflow/issues/8682), [#8683](https://togithub.com/mlflow/mlflow/issues/8683), [@dbczumar](https://togithub.com/dbczumar); [#8791](https://togithub.com/mlflow/mlflow/issues/8791), [@mhattingpete](https://togithub.com/mhattingpete); [#8739](https://togithub.com/mlflow/mlflow/issues/8739), [@yunpark93](https://togithub.com/yunpark93)
### [`v2.4.2`](https://togithub.com/mlflow/mlflow/blob/HEAD/CHANGELOG.md#242-2023-07-10)
[Compare Source](https://togithub.com/mlflow/mlflow/compare/v2.4.1...v2.4.2)
MLflow 2.4.2 is a patch release containing the following bug fixes and changes:
Bug fixes:
- \[Models] Add compatibility for legacy transformers serialization ([#8964](https://togithub.com/mlflow/mlflow/issues/8964), [@BenWilson2](https://togithub.com/BenWilson2))
- \[Models] Fix downloading MLmodel files from alias-based models:/ URIs ([#8764](https://togithub.com/mlflow/mlflow/issues/8764), [@smurching](https://togithub.com/smurching))
- \[Models] Fix reading model flavor config from URI for models in UC ([#8728](https://togithub.com/mlflow/mlflow/issues/8728), [@smurching](https://togithub.com/smurching))
- \[Models] Support `feature_deps` in ModelVersion creation for UC ([#8867](https://togithub.com/mlflow/mlflow/issues/8867), [#8815](https://togithub.com/mlflow/mlflow/issues/8815), [@AveshCSingh](https://togithub.com/AveshCSingh))
- \[Models] Add support for listing artifacts in UC model registry artifact repo ([#8803](https://togithub.com/mlflow/mlflow/issues/8803), [@smurching](https://togithub.com/smurching))
- \[Core] Include resources for recipes in mlflow-skinny ([#8895](https://togithub.com/mlflow/mlflow/issues/8895), [@harupy](https://togithub.com/harupy))
- \[UI] Enable datasets tracking UI ([#8886](https://togithub.com/mlflow/mlflow/issues/8886), [@harupy](https://togithub.com/harupy))
- \[Artifacts] Use `MLFLOW_ENABLE_MULTIPART_DOWNLOAD` in `DatabricksArtifactRepository` ([#8884](https://togithub.com/mlflow/mlflow/issues/8884), [@harupy](https://togithub.com/harupy))
Documentation updates:
- \[Examples / Docs] Add question-answering and summarization examples and docs with LLMs ([#8695](https://togithub.com/mlflow/mlflow/issues/8695), [@dbczumar](https://togithub.com/dbczumar))
- \[Examples / Docs] Add johnsnowlabs flavor example and doc ([#8689](https://togithub.com/mlflow/mlflow/issues/8689), [@C-K-Loan](https://togithub.com/C-K-Loan))
Small bug fixes and documentation updates:
[#8966](https://togithub.com/mlflow/mlflow/issues/8966), [@BenWilson2](https://togithub.com/BenWilson2); [#8881](https://togithub.com/mlflow/mlflow/issues/8881), [@harupy](https://togithub.com/harupy); [#8846](https://togithub.com/mlflow/mlflow/issues/8846), [#8760](https://togithub.com/mlflow/mlflow/issues/8760), [@smurching](https://togithub.com/smurching)
### [`v2.4.1`](https://togithub.com/mlflow/mlflow/blob/HEAD/CHANGELOG.md#241-2023-06-09)
[Compare Source](https://togithub.com/mlflow/mlflow/compare/v2.4.0...v2.4.1)
MLflow 2.4.1 is a patch release containing the following features, bug fixes and changes:
Features:
- \[Tracking] Extend SearchRuns to support datasets ([#8622](https://togithub.com/mlflow/mlflow/issues/8622), [@prithvikannan](https://togithub.com/prithvikannan))
- \[Models] Add an `mlflow.johnsnowlabs` flavor for the `johnsnowlabs` package ([#8556](https://togithub.com/mlflow/mlflow/issues/8556), [@C-K-Loan](https://togithub.com/C-K-Loan))
- \[Models] Add a warning for duplicate pip requirements specified in `save_model` and `log_model` for the `transformers` flavor ([#8678](https://togithub.com/mlflow/mlflow/issues/8678), [@BenWilson2](https://togithub.com/BenWilson2))
Bug fixes:
- \[Security] Improve robustness to LFI attacks ([#8648](https://togithub.com/mlflow/mlflow/issues/8648), [@serena-ruan](https://togithub.com/serena-ruan))
- If you are using `mlflow server` or `mlflow ui`, we recommend upgrading to MLflow 2.4.1 as soon as possible.
- \[Models] Fix an issue with `transformers` serialization for ModelCards that contain invalid characters ([#8652](https://togithub.com/mlflow/mlflow/issues/8652), [@BenWilson2](https://togithub.com/BenWilson2))
- \[Models] Fix connection pooling deadlocks that occurred during large file downloads ([#8682](https://togithub.com/mlflow/mlflow/issues/8682), [@dbczumar](https://togithub.com/dbczumar); [#8660](https://togithub.com/mlflow/mlflow/issues/8660), [@harupy](https://togithub.com/harupy))
Small bug fixes and documentation updates:
[#8677](https://togithub.com/mlflow/mlflow/issues/8677), [#8674](https://togithub.com/mlflow/mlflow/issues/8674), [#8646](https://togithub.com/mlflow/mlflow/issues/8646), [#8647](https://togithub.com/mlflow/mlflow/issues/8647), [@dbczumar](https://togithub.com/dbczumar); [#8654](https://togithub.com/mlflow/mlflow/issues/8654), [#8653](https://togithub.com/mlflow/mlflow/issues/8653), [#8660](https://togithub.com/mlflow/mlflow/issues/8660), [#8650](https://togithub.com/mlflow/mlflow/issues/8650), [#8642](https://togithub.com/mlflow/mlflow/issues/8642), [#8636](https://togithub.com/mlflow/mlflow/issues/8636), [#8599](https://togithub.com/mlflow/mlflow/issues/8599), [#8637](https://togithub.com/mlflow/mlflow/issues/8637), [#8608](https://togithub.com/mlflow/mlflow/issues/8608), [#8633](https://togithub.com/mlflow/mlflow/issues/8633), [#8623](https://togithub.com/mlflow/mlflow/issues/8623), [#8628](https://togithub.com/mlflow/mlflow/issues/8628), [#8619](https://togithub.com/mlflow/mlflow/issues/8619), [@harupy](https://togithub.com/harupy); [#8655](https://togithub.com/mlflow/mlflow/issues/8655), [#8609](https://togithub.com/mlflow/mlflow/issues/8609), [@BenWilson2](https://togithub.com/BenWilson2); [#8648](https://togithub.com/mlflow/mlflow/issues/8648), [@serena-ruan](https://togithub.com/serena-ruan); [#8521](https://togithub.com/mlflow/mlflow/issues/8521), [@ka1mar](https://togithub.com/ka1mar); [#8638](https://togithub.com/mlflow/mlflow/issues/8638), [@smurching](https://togithub.com/smurching); [#8634](https://togithub.com/mlflow/mlflow/issues/8634), [@PenHsuanWang](https://togithub.com/PenHsuanWang)
### [`v2.4.0`](https://togithub.com/mlflow/mlflow/blob/HEAD/CHANGELOG.md#240-2023-06-06)
[Compare Source](https://togithub.com/mlflow/mlflow/compare/v2.3.2...v2.4.0)
MLflow 2.4.0 includes several major features and improvements
Features:
- \[Tracking] Introduce dataset tracking APIs: `mlflow.data` and `mlflow.log_input()` ([#8186](https://togithub.com/mlflow/mlflow/issues/8186), [@prithvikannan](https://togithub.com/prithvikannan))
- \[Tracking] Add `mlflow.log_table()` and `mlflow.load_table()` APIs for logging evaluation tables ([#8523](https://togithub.com/mlflow/mlflow/issues/8523), [#8467](https://togithub.com/mlflow/mlflow/issues/8467), [@sunishsheth2009](https://togithub.com/sunishsheth2009))
- \[Tracking] Introduce `mlflow.get_parent_run()` fluent API ([#8493](https://togithub.com/mlflow/mlflow/issues/8493), [@annzhang-db](https://togithub.com/annzhang-db))
- \[Tracking / Model Registry] Re-introduce faster artifact downloads on Databricks ([#8352](https://togithub.com/mlflow/mlflow/issues/8352), [@dbczumar](https://togithub.com/dbczumar); [#8561](https://togithub.com/mlflow/mlflow/issues/8561), [@harupy](https://togithub.com/harupy))
- \[UI] Add dataset tracking information to MLflow Tracking UI ([#8602](https://togithub.com/mlflow/mlflow/issues/8602), [@prithvikannan](https://togithub.com/prithvikannan), [@hubertzub-db](https://togithub.com/hubertzub-db))
- \[UI] Introduce Artifact View for comparing inputs, outputs, and metadata across models ([#8602](https://togithub.com/mlflow/mlflow/issues/8602), [@hubertzub-db](https://togithub.com/hubertzub-db))
- \[Models] Extend `mlflow.evaluate()` to support LLM tasks ([#8484](https://togithub.com/mlflow/mlflow/issues/8484), [@harupy](https://togithub.com/harupy))
- \[Models] Support logging subclasses of `Chain` and `LLMChain` in `mlflow.langchain` flavor ([#8453](https://togithub.com/mlflow/mlflow/issues/8453), [@liangz1](https://togithub.com/liangz1))
- \[Models] Add support for LangChain Agents to the `mlflow.langchain` flavor ([#8297](https://togithub.com/mlflow/mlflow/issues/8297), [@sunishsheth2009](https://togithub.com/sunishsheth2009))
- \[Models] Add a `mlflow.sentence_transformers` flavor for SentenceTransformers ([#8479](https://togithub.com/mlflow/mlflow/issues/8479), [@BenWilson2](https://togithub.com/BenWilson2); [#8547](https://togithub.com/mlflow/mlflow/issues/8547), [@Loquats](https://togithub.com/Loquats))
- \[Models] Add support for multi-GPU inference and efficient weight loading for `mlflow.transformers` flavor ([#8448](https://togithub.com/mlflow/mlflow/issues/8448), [@ankit-db](https://togithub.com/ankit-db))
- \[Models] Support the `max_shard_size` parameter in the `mlflow.transformers` flavor ([#8567](https://togithub.com/mlflow/mlflow/issues/8567), [@wenfeiy-db](https://togithub.com/wenfeiy-db))
- \[Models] Add support for audio transcription pipelines in the `mlflow.transformers` flavor ([#8464](https://togithub.com/mlflow/mlflow/issues/8464), [@BenWilson2](https://togithub.com/BenWilson2))
- \[Models] Add support for audio classification to `mlflow.transformers` flavor ([#8492](https://togithub.com/mlflow/mlflow/issues/8492), [@BenWilson2](https://togithub.com/BenWilson2))
- \[Models] Add support for URI inputs in audio models logged with the `mlflow.transformers` flavor ([#8495](https://togithub.com/mlflow/mlflow/issues/8495), [@BenWilson2](https://togithub.com/BenWilson2))
- \[Models] Add support for returning classifier scores in `mlflow.transformers` pyfunc outputs ([#8512](https://togithub.com/mlflow/mlflow/issues/8512), [@BenWilson2](https://togithub.com/BenWilson2))
- \[Models] Support optional inputs in model signatures ([#8438](https://togithub.com/mlflow/mlflow/issues/8438), [@apurva-koti](https://togithub.com/apurva-koti))
- \[Models] Introduce an `mlflow.models.set_signature()` API to set the signature of a logged model ([#8476](https://togithub.com/mlflow/mlflow/issues/8476), [@jerrylian-db](https://togithub.com/jerrylian-db))
- \[Models] Persist ONNX Runtime InferenceSession options when logging a model with `mlflow.onnx.log_model()` ([#8433](https://togithub.com/mlflow/mlflow/issues/8433), [@leqiao-1](https://togithub.com/leqiao-1))
Bug fixes:
- \[Tracking] Terminate Spark callback server when Spark Autologging is disabled or Spark Session is shut down ([#8508](https://togithub.com/mlflow/mlflow/issues/8508), [@WeichenXu123](https://togithub.com/WeichenXu123))
- \[Tracking] Fix compatibility of `mlflow server` with `Flask<2.0` ([#8463](https://togithub.com/mlflow/mlflow/issues/8463), [@kevingreer](https://togithub.com/kevingreer))
- \[Models] Convert `mlflow.transformers` pyfunc scalar string output to list of strings during batch inference ([#8546](https://togithub.com/mlflow/mlflow/issues/8546), [@BenWilson2](https://togithub.com/BenWilson2))
- \[Models] Fix a bug causing outdated pyenv versions to be installed by `mlflow models build-docker` ([#8488](https://togithub.com/mlflow/mlflow/issues/8488), [@Hellzed](https://togithub.com/Hellzed))
- \[Model Registry] Remove aliases from storage when a Model Version is deleted ([#8459](https://togithub.com/mlflow/mlflow/issues/8459), [@arpitjasa-db](https://togithub.com/arpitjasa-db))
Documentation updates:
- \[Docs] Publish a new MLOps Quickstart for model selection and deployment ([#8462](https://togithub.com/mlflow/mlflow/issues/8462), [@lobrien](https://togithub.com/lobrien))
- \[Docs] Add MLflavors library to Community Model Flavors documentation ([#8420](https://togithub.com/mlflow/mlflow/issues/8420), [@benjaminbluhm](https://togithub.com/benjaminbluhm))
- \[Docs] Add documentation for Registered Model Aliases ([#8445](https://togithub.com/mlflow/mlflow/issues/8445), [@arpitjasa-db](https://togithub.com/arpitjasa-db))
- \[Docs] Fix errors in documented `mlflow models` CLI command examples ([#8480](https://togithub.com/mlflow/mlflow/issues/8480), [@vijethmoudgalya](https://togithub.com/vijethmoudgalya))
Small bug fixes and documentation updates:
[#8611](https://togithub.com/mlflow/mlflow/issues/8611), [#8587](https://togithub.com/mlflow/mlflow/issues/8587), [@dbczumar](https://togithub.com/dbczumar); [#8617](https://togithub.com/mlflow/mlflow/issues/8617), [#8620](https://togithub.com/mlflow/mlflow/issues/8620), [#8615](https://togithub.com/mlflow/mlflow/issues/8615), [#8603](https://togithub.com/mlflow/mlflow/issues/8603), [#8604](https://togithub.com/mlflow/mlflow/issues/8604), [#8601](https://togithub.com/mlflow/mlflow/issues/8601), [#8596](https://togithub.com/mlflow/mlflow/issues/8596), [#8598](https://togithub.com/mlflow/mlflow/issues/8598), [#8597](https://togithub.com/mlflow/mlflow/issues/8597), [#8589](https://togithub.com/mlflow/mlflow/issues/8589), [#8580](https://togithub.com/mlflow/mlflow/issues/8580), [#8581](https://togithub.com/mlflow/mlflow/issues/8581), [#8575](https://togithub.com/mlflow/mlflow/issues/8575), [#8582](https://togithub.com/mlflow/mlflow/issues/8582), [#8577](https://togithub.com/mlflow/mlflow/issues/8577), [#8576](https://togithub.com/mlflow/mlflow/issues/8576), [#8578](https://togithub.com/mlflow/mlflow/issues/8578), [#8561](https://togithub.com/mlflow/mlflow/issues/8561), [#8568](https://togithub.com/mlflow/mlflow/issues/8568), [#8551](https://togithub.com/mlflow/mlflow/issues/8551), [#8528](https://togithub.com/mlflow/mlflow/issues/8528), [#8550](https://togithub.com/mlflow/mlflow/issues/8550), [#8489](https://togithub.com/mlflow/mlflow/issues/8489), [#8530](https://togithub.com/mlflow/mlflow/issues/8530), [#8534](https://togithub.com/mlflow/mlflow/issues/8534), [#8533](https://togithub.com/mlflow/mlflow/issues/8533), [#8532](https://togithub.com/mlflow/mlflow/issues/8532), [#8524](https://togithub.com/mlflow/mlflow/issues/8524), [#8520](https://togithub.com/mlflow/mlflow/issues/8520), [#8517](https://togithub.com/mlflow/mlflow/issues/8517), [#8516](https://togithub.com/mlflow/mlflow/issues/8516), [#8515](https://togithub.com/mlflow/mlflow/issues/8515), [#8514](https://togithub.com/mlflow/mlflow/issues/8514), [#8506](https://togithub.com/mlflow/mlflow/issues/8506), [#8503](https://togithub.com/mlflow/mlflow/issues/8503), [#8500](https://togithub.com/mlflow/mlflow/issues/8500), [#8504](https://togithub.com/mlflow/mlflow/issues/8504), [#8496](https://togithub.com/mlflow/mlflow/issues/8496), [#8486](https://togithub.com/mlflow/mlflow/issues/8486), [#8485](https://togithub.com/mlflow/mlflow/issues/8485), [#8468](https://togithub.com/mlflow/mlflow/issues/8468), [#8471](https://togithub.com/mlflow/mlflow/issues/8471), [#8473](https://togithub.com/mlflow/mlflow/issues/8473), [#8470](https://togithub.com/mlflow/mlflow/issues/8470), [#8458](https://togithub.com/mlflow/mlflow/issues/8458), [#8447](https://togithub.com/mlflow/mlflow/issues/8447), [#8446](https://togithub.com/mlflow/mlflow/issues/8446), [#8434](https://togithub.com/mlflow/mlflow/issues/8434), [@harupy](https://togithub.com/harupy); [#8607](https://togithub.com/mlflow/mlflow/issues/8607), [#8538](https://togithub.com/mlflow/mlflow/issues/8538), [#8513](https://togithub.com/mlflow/mlflow/issues/8513), [#8452](https://togithub.com/mlflow/mlflow/issues/8452), [#8466](https://togithub.com/mlflow/mlflow/issues/8466), [#8465](https://togithub.com/mlflow/mlflow/issues/8465), [@serena-ruan](https://togithub.com/serena-ruan); [#8586](https://togithub.com/mlflow/mlflow/issues/8586), [#8595](https://togithub.com/mlflow/mlflow/issues/8595), [@prithvikannan](https://togithub.com/prithvikannan); [#8593](https://togithub.com/mlflow/mlflow/issues/8593), [#8541](https://togithub.com/mlflow/mlflow/issues/8541), [@kriscon-db](https://togithub.com/kriscon-db); [#8592](https://togithub.com/mlflow/mlflow/issues/8592), [#8566](https://togithub.com/mlflow/mlflow/issues/8566), [@annzhang-db](https://togithub.com/annzhang-db); [#8588](https://togithub.com/mlflow/mlflow/issues/8588), [#8565](https://togithub.com/mlflow/mlflow/issues/8565), [#8559](https://togithub.com/mlflow/mlflow/issues/8559), [#8537](https://togithub.com/mlflow/mlflow/issues/8537), [@BenWilson2](https://togithub.com/BenWilson2); [#8545](https://togithub.com/mlflow/mlflow/issues/8545), [@apurva-koti](https://togithub.com/apurva-koti); [#8564](https://togithub.com/mlflow/mlflow/issues/8564), [@DavidSpek](https://togithub.com/DavidSpek); [#8436](https://togithub.com/mlflow/mlflow/issues/8436), [#8490](https://togithub.com/mlflow/mlflow/issues/8490), [@jerrylian-db](https://togithub.com/jerrylian-db); [#8505](https://togithub.com/mlflow/mlflow/issues/8505), [@eliaskoromilas](https://togithub.com/eliaskoromilas); [#8483](https://togithub.com/mlflow/mlflow/issues/8483), [@WeichenXu123](https://togithub.com/WeichenXu123); [#8472](https://togithub.com/mlflow/mlflow/issues/8472), [@leqiao-1](https://togithub.com/leqiao-1); [#8429](https://togithub.com/mlflow/mlflow/issues/8429), [@jinzhang21](https://togithub.com/jinzhang21); [#8581](https://togithub.com/mlflow/mlflow/issues/8581), [#8548](https://togithub.com/mlflow/mlflow/issues/8548), [#8499](https://togithub.com/mlflow/mlflow/issues/8499), [@gabrielfu](https://togithub.com/gabrielfu);
### [`v2.3.2`](https://togithub.com/mlflow/mlflow/blob/HEAD/CHANGELOG.md#232-2023-05-12)
[Compare Source](https://togithub.com/mlflow/mlflow/compare/v2.3.1...v2.3.2)
MLflow 2.3.2 is a patch release containing the following features, bug fixes and changes:
Features:
- \[Models] Add GPU support for `transformers` models `pyfunc` inference and serving ([#8375](https://togithub.com/mlflow/mlflow/issues/8375), [@ankit-db](https://togithub.com/ankit-db))
- \[Models] Disable autologging functionality for non-relevant models when training a `transformers` model ([#8405](https://togithub.com/mlflow/mlflow/issues/8405), [@BenWilson2](https://togithub.com/BenWilson2))
- \[Models] Add support for preserving and overriding `torch_dtype` values in `transformers` pipelines ([#8421](https://togithub.com/mlflow/mlflow/issues/8421), [@BenWilson2](https://togithub.com/BenWilson2))
- \[Models] Add support for `Feature Extraction` pipelines in the `transformers` flavor ([#8423](https://togithub.com/mlflow/mlflow/issues/8423), [@BenWilson2](https://togithub.com/BenWilson2))
- \[Tracking] Add basic HTTP auth support for users, registered models, and experiments permissions ([#8286](https://togithub.com/mlflow/mlflow/issues/8286), [@gabrielfu](https://togithub.com/gabrielfu))
Bug Fixes:
- \[Models] Fix inferred schema issue with `Text2TextGeneration` pipelines in the `transformers` flavor ([#8391](https://togithub.com/mlflow/mlflow/issues/8391), [@BenWilson2](https://togithub.com/BenWilson2))
- \[Models] Change MLflow dependency pinning in logged models from a range value to an exact major and minor version ([#8422](https://togithub.com/mlflow/mlflow/issues/8422), [@harupy](https://togithub.com/harupy))
Documentation updates:
- \[Examples] Add `signature` logging to all examples and documentation ([#8410](https://togithub.com/mlflow/mlflow/issues/8410), [#8401](https://togithub.com/mlflow/mlflow/issues/8401), [#8400](https://togithub.com/mlflow/mlflow/issues/8400), [#8387](https://togithub.com/mlflow/mlflow/issues/8387) [@jerrylian-db](https://togithub.com/jerrylian-db))
- \[Examples] Add `sentence-transformers` examples to the `transformers` examples suite ([#8425](https://togithub.com/mlflow/mlflow/issues/8425), [@BenWilson2](https://togithub.com/BenWilson2))
- \[Docs] Add a new MLflow Quickstart documentation page ([#8171](https://togithub.com/mlflow/mlflow/issues/8171), [@lobrien](https://togithub.com/lobrien))
- \[Docs] Add a new introduction to MLflow page ([#8365](https://togithub.com/mlflow/mlflow/issues/8365), [@lobrien](https://togithub.com/lobrien))
- \[Docs] Add a community model plugin example and documentation for `trubrics` ([#8371](https://togithub.com/mlflow/mlflow/issues/8371), [@jeffkayne](https://togithub.com/jeffkayne))
- \[Docs] Add `gluon` pyfunc example to Model flavor documentation ([#8403](https://togithub.com/mlflow/mlflow/issues/8403), [@ericvincent18](https://togithub.com/ericvincent18))
- \[Docs] Add `statsmodels` pyfunc example to `Models` flavor documentation ([#8394](https://togithub.com/mlflow/mlflow/issues/8394), [@ericvincent18](https://togithub.com/ericvincent18))
Small bug fixes and documentation updates:
[#8415](https://togithub.com/mlflow/mlflow/issues/8415), [#8412](https://togithub.com/mlflow/mlflow/issues/8412), [#8411](https://togithub.com/mlflow/mlflow/issues/8411), [#8355](https://togithub.com/mlflow/mlflow/issues/8355), [#8354](https://togithub.com/mlflow/mlflow/issues/8354), [#8353](https://togithub.com/mlflow/mlflow/issues/8353), [#8348](https://togithub.com/mlflow/mlflow/issues/8348), [@harupy](https://togithub.com/harupy); [#8374](https://togithub.com/mlflow/mlflow/issues/8374), [#8367](https://togithub.com/mlflow/mlflow/issues/8367), [#8350](https://togithub.com/mlflow/mlflow/issues/8350), [@dbczumar](https://togithub.com/dbczumar); [#8358](https://togithub.com/mlflow/mlflow/issues/8358) [@mrkaye97](https://togithub.com/mrkaye97); [#8392](https://togithub.com/mlflow/mlflow/issues/8392), [#8362](https://togithub.com/mlflow/mlflow/issues/8362), [@smurching](https://togithub.com/smurching); [#8427](https://togithub.com/mlflow/mlflow/issues/8427), [#8408](https://togithub.com/mlflow/mlflow/issues/8408), [#8399](https://togithub.com/mlflow/mlflow/issues/8399), [#8381](https://togithub.com/mlflow/mlflow/issues/8381), [@BenWilson2](https://togithub.com/BenWilson2); [#8395](https://togithub.com/mlflow/mlflow/issues/8395), [#8390](https://togithub.com/mlflow/mlflow/issues/8390), [@jerrylian-db](https://togithub.com/jerrylian-db); [#8402](https://togithub.com/mlflow/mlflow/issues/8402), [#8398](https://togithub.com/mlflow/mlflow/issues/8398), [@WeichenXu123](https://togithub.com/WeichenXu123); [#8377](https://togithub.com/mlflow/mlflow/issues/8377), [#8363](https://togithub.com/mlflow/mlflow/issues/8363), [@arpitjasa-db](https://togithub.com/arpitjasa-db); [#8385](https://togithub.com/mlflow/mlflow/issues/8385), [@prithvikannan](https://togithub.com/prithvikannan); [#8418](https://togithub.com/mlflow/mlflow/issues/8418), [@Jeukoh](https://togithub.com/Jeukoh);
### [`v2.3.1`](https://togithub.com/mlflow/mlflow/blob/HEAD/CHANGELOG.md#231-2023-04-27)
[Compare Source](https://togithub.com/mlflow/mlflow/compare/v2.3.0...v2.3.1)
MLflow 2.3.1 is a patch release containing the following bug fixes and changes:
Bug fixes:
- \[Security] Fix critical LFI attack vulnerability by disabling the ability to provide relative paths in registered model sources ([#8281](https://togithub.com/mlflow/mlflow/issues/8281), [@BenWilson2](https://togithub.com/BenWilson2))
- **If you are using `mlflow server` or `mlflow ui`, we recommend upgrading to MLflow 2.3.1 as soon as possible.** For more details, see https://github.com/mlflow/mlflow/security/advisories/GHSA-xg73-94fp-g449.
- \[Tracking] Fix an issue causing file and model uploads to hang on Databricks ([#8348](https://togithub.com/mlflow/mlflow/issues/8348), [@harupy](https://togithub.com/harupy))
- \[Tracking / Model Registry] Fix an issue causing file and model downloads to hang on Databricks ([#8350](https://togithub.com/mlflow/mlflow/issues/8350), [@dbczumar](https://togithub.com/dbczumar))
- \[Scoring] Fix regression in schema enforcement for model serving when using the `inputs` format for inference ([#8326](https://togithub.com/mlflow/mlflow/issues/8326), [@BenWilson2](https://togithub.com/BenWilson2))
- \[Model Registry] Fix regression in model naming parsing where special characters were not accepted in model names ([#8322](https://togithub.com/mlflow/mlflow/issues/8322), [@arpitjasa-db](https://togithub.com/arpitjasa-db))
- \[Recipes] Fix card rendering with the pandas profiler to handle columns containing all null values ([#8263](https://togithub.com/mlflow/mlflow/issues/8263), [@sunishsheth2009](https://togithub.com/sunishsheth2009))
Documentation updates:
- \[Docs] Add an H2O pyfunc usage example to the models documentation ([#8292](https://togithub.com/mlflow/mlflow/issues/8292), [@ericvincent18](https://togithub.com/ericvincent18))
- \[Examples] Add a TensorFlow Core 2.x API usage example ([#8235](https://togithub.com/mlflow/mlflow/issues/8235), [@dheerajnbhat](https://togithub.com/dheerajnbhat))
Small bug fixes and documentation updates:
[#8324](https://togithub.com/mlflow/mlflow/issues/8324), [#8325](https://togithub.com/mlflow/mlflow/issues/8325), [@smurching](https://togithub.com/smurching); [#8313](https://togithub.com/mlflow/mlflow/issues/8313), [@dipanjank](https://togithub.com/dipanjank); [#8323](https://togithub.com/mlflow/mlflow/issues/8323), [@liangz1](https://togithub.com/liangz1); [#8331](https://togithub.com/mlflow/mlflow/issues/8331), [#8328](https://togithub.com/mlflow/mlflow/issues/8328), [#8319](https://togithub.com/mlflow/mlflow/issues/8319), [#8316](https://togithub.com/mlflow/mlflow/issues/8316), [#8308](https://togithub.com/mlflow/mlflow/issues/8308), [#8293](https://togithub.com/mlflow/mlflow/issues/8293), [#8289](https://togithub.com/mlflow/mlflow/issues/8289), [#8283](https://togithub.com/mlflow/mlflow/issues/8283), [#8284](https://togithub.com/mlflow/mlflow/issues/8284), [#8285](https://togithub.com/mlflow/mlflow/issues/8285), [#8282](https://togithub.com/mlflow/mlflow/issues/8282), [#8241](https://togithub.com/mlflow/mlflow/issues/8241), [#8270](https://togithub.com/mlflow/mlflow/issues/8270), [#8272](https://togithub.com/mlflow/mlflow/issues/8272), [#8271](https://togithub.com/mlflow/mlflow/issues/8271), [#8268](https://togithub.com/mlflow/mlflow/issues/8268), [@harupy](https://togithub.com/harupy); [#8312](https://togithub.com/mlflow/mlflow/issues/8312), [#8294](https://togithub.com/mlflow/mlflow/issues/8294), [#8295](https://togithub.com/mlflow/mlflow/issues/8295), [#8279](https://togithub.com/mlflow/mlflow/issues/8279), [#8267](https://togithub.com/mlflow/mlflow/issues/8267), [@BenWilson2](https://togithub.com/BenWilson2); [#8290](https://togithub.com/mlflow/mlflow/issues/8290), [@jinzhang21](https://togithub.com/jinzhang21); [#8257](https://togithub.com/mlflow/mlflow/issues/8257), [@WeichenXu123](https://togithub.com/WeichenXu123); [#8307](https://togithub.com/mlflow/mlflow/issues/8307), [@arpitjasa-db](https://togithub.com/arpitjasa-db)
### [`v2.3.0`](https://togithub.com/mlflow/mlflow/blob/HEAD/CHANGELOG.md#230-2023-04-18)
[Compare Source](https://togithub.com/mlflow/mlflow/compare/v2.2.2...v2.3.0)
MLflow 2.3.0 includes several major features and improvements
Features:
- \[Models] Introduce a new `transformers` named flavor ([#8236](https://togithub.com/mlflow/mlflow/issues/8236), [#8181](https://togithub.com/mlflow/mlflow/issues/8181), [#8086](https://togithub.com/mlflow/mlflow/issues/8086), [@BenWilson2](https://togithub.com/BenWilson2))
- \[Models] Introduce a new `openai` named flavor ([#8191](https://togithub.com/mlflow/mlflow/issues/8191), [#8155](https://togithub.com/mlflow/mlflow/issues/8155), [@harupy](https://togithub.com/harupy))
- \[Models] Introduce a new `langchain` named flavor ([#8251](https://togithub.com/mlflow/mlflow/issues/8251), [#8197](https://togithub.com/mlflow/mlflow/issues/8197), [@liangz1](https://togithub.com/liangz1), [@sunishsheth2009](https://togithub.com/sunishsheth2009))
- \[Models] Add support for `Pytorch` and `Lightning` 2.0 ([#8072](https://togithub.com/mlflow/mlflow/issues/8072), [@shrinath-suresh](https://togithub.com/shrinath-suresh))
- \[Tracking] Add support for logging LLM input, output, and prompt artifacts ([#8234](https://togithub.com/mlflow/mlflow/issues/8234), [#8204](https://togithub.com/mlflow/mlflow/issues/8204), [@sunishsheth2009](https://togithub.com/sunishsheth2009))
- \[Tracking] Add support for HTTP Basic Auth in the MLflow tracking server ([#8130](https://togithub.com/mlflow/mlflow/issues/8130), [@gabrielfu](https://togithub.com/gabrielfu))
- \[Tracking] Add `search_model_versions` to the fluent API ([#8223](https://togithub.com/mlflow/mlflow/issues/8223), [@mariusschlegel](https://togithub.com/mariusschlegel))
- \[Artifacts] Add support for parallelized artifact downloads ([#8116](https://togithub.com/mlflow/mlflow/issues/8116), [@apurva-koti](https://togithub.com/apurva-koti))
- \[Artifacts] Add support for parallelized artifact uploads for AWS ([#8003](https://togithub.com/mlflow/mlflow/issues/8003), [@harupy](https://togithub.com/harupy))
- \[Artifacts] Add content type headers to artifact upload requests for the `HttpArtifactRepository` ([#8048](https://togithub.com/mlflow/mlflow/issues/8048), [@WillEngler](https://togithub.com/WillEngler))
- \[Model Registry] Add alias support for logged models within Model Registry ([#8164](https://togithub.com/mlflow/mlflow/issues/8164), [#8094](https://togithub.com/mlflow/mlflow/issues/8094), [#8055](https://togithub.com/mlflow/mlflow/issues/8055) [@arpitjasa-db](https://togithub.com/arpitjasa-db))
- \[UI] Add support for custom domain git providers ([#7933](https://togithub.com/mlflow/mlflow/issues/7933), [@gusghrlrl101](https://togithub.com/gusghrlrl101))
- \[Scoring] Add plugin support for customization of MLflow serving endpoints ([#7757](https://togithub.com/mlflow/mlflow/issues/7757), [@jmahlik](https://togithub.com/jmahlik))
- \[Scoring] Add support to MLflow serving that allows configuration of multiple inference workers ([#8035](https://togithub.com/mlflow/mlflow/issues/8035), [@M4nouel](https://togithub.com/M4nouel))
- \[Sagemaker] Add support for asynchronous inference configuration on Sagemaker ([#8009](https://togithub.com/mlflow/mlflow/issues/8009), [@thomasbell1985](https://togithub.com/thomasbell1985))
- \[Build] Remove `shap` as a core dependency of MLflow ([#8199](https://togithub.com/mlflow/mlflow/issues/8199), [@jmahlik](https://togithub.com/jmahlik))
Bug fixes:
- \[Models] Fix a bug with `tensorflow` autologging for models with multiple inputs ([#8097](https://togithub.com/mlflow/mlflow/issues/8097), [@jaume-ferrarons](https://togithub.com/jaume-ferrarons))
- \[Recipes] Fix a bug with `Pandas` 2.0 updates for profiler rendering of datetime types ([#7925](https://togithub.com/mlflow/mlflow/issues/7925), [@sunishsheth2009](https://togithub.com/sunishsheth2009))
- \[Tracking] Prevent exceptions from being raised if a parameter is logged with an existing key whose value is identical to the logged parameter ([#8038](https://togithub.com/mlflow/mlflow/issues/8038), [@AdamStelmaszczyk](https://togithub.com/AdamStelmaszczyk))
- \[Tracking] Fix an issue with deleting experiments in the FileStore backend ([#8178](https://togithub.com/mlflow/
Configuration
📅 Schedule: Branch creation - "" (UTC), Automerge - At any time (no schedule defined).
🚦 Automerge: Disabled by config. Please merge this manually once you are satisfied.
♻ Rebasing: Whenever PR becomes conflicted, or you tick the rebase/retry checkbox.
🔕 Ignore: Close this PR and you won't be reminded about these updates again.
[ ] If you want to rebase/retry this PR, check this box
This PR has been generated by Mend Renovate. View repository job log here.
This PR contains the following updates:
==1.26.1->==2.5.0==1.25.1->==2.5.0GitHub Vulnerability Alerts
CVE-2023-1176
Impact
Users of the MLflow Open Source Project who are hosting the MLflow Model Registry using the
mlflow serverormlflow uicommands using an MLflow version older than MLflow 2.2.1 may be vulnerable to a remote file existence check exploit if they are not limiting who can query their server (for example, by using a cloud VPC, an IP allowlist for inbound requests, or authentication / authorization middleware).This issue only affects users and integrations that run the
mlflow serverandmlflow uicommands. Integrations that do not make use ofmlflow serverormlflow uiare unaffected; for example, the Databricks Managed MLflow product and MLflow on Azure Machine Learning do not make use of these commands and are not impacted by these vulnerabilities in any way.The vulnerability detailed in https://nvd.nist.gov/vuln/detail/CVE-2023-1176 enables an actor to check the existence of arbitrary files unrelated to MLflow from the host server, including any files stored in remote locations to which the host server has access.
Patches
This vulnerability has been patched in MLflow 2.2.1, which was released to PyPI on March 2nd, 2023. If you are using
mlflow serverormlflow uiwith the MLflow Model Registry, we recommend upgrading to MLflow 2.2.1 as soon as possible.Workarounds
If you are using the MLflow open source
mlflow serverormlflow uicommands, we strongly recommend limiting who can access your MLflow Model Registry and MLflow Tracking servers using a cloud VPC, an IP allowlist for inbound requests, authentication / authorization middleware, or another access restriction mechanism of your choosing.If you are using the MLflow open source
mlflow serverormlflow uicommands, we also strongly recommend limiting the remote files to which your MLflow Model Registry and MLflow Tracking servers have access. For example, if your MLflow Model Registry or MLflow Tracking server uses cloud-hosted blob storage for MLflow artifacts, make sure to restrict the scope of your server's cloud credentials such that it can only access files and directories related to MLflow.References
More information about the vulnerability is available at https://nvd.nist.gov/vuln/detail/CVE-2023-1176.
CVE-2023-1177
Impact
Users of the MLflow Open Source Project who are hosting the MLflow Model Registry using the
mlflow serverormlflow uicommands using an MLflow version older than MLflow 2.2.1 may be vulnerable to a remote file access exploit if they are not limiting who can query their server (for example, by using a cloud VPC, an IP allowlist for inbound requests, or authentication / authorization middleware).This issue only affects users and integrations that run the
mlflow serverandmlflow uicommands. Integrations that do not make use ofmlflow serverormlflow uiare unaffected; for example, the Databricks Managed MLflow product and MLflow on Azure Machine Learning do not make use of these commands and are not impacted by these vulnerabilities in any way.The vulnerability detailed in https://nvd.nist.gov/vuln/detail/CVE-2023-1177 enables an actor to download arbitrary files unrelated to MLflow from the host server, including any files stored in remote locations to which the host server has access.
Patches
This vulnerability has been patched in MLflow 2.2.1, which was released to PyPI on March 2nd, 2023. If you are using
mlflow serverormlflow uiwith the MLflow Model Registry, we recommend upgrading to MLflow 2.2.1 as soon as possible.Workarounds
If you are using the MLflow open source
mlflow serverormlflow uicommands, we strongly recommend limiting who can access your MLflow Model Registry and MLflow Tracking servers using a cloud VPC, an IP allowlist for inbound requests, authentication / authorization middleware, or another access restriction mechanism of your choosing.If you are using the MLflow open source
mlflow serverormlflow uicommands, we also strongly recommend limiting the remote files to which your MLflow Model Registry and MLflow Tracking servers have access. For example, if your MLflow Model Registry or MLflow Tracking server uses cloud-hosted blob storage for MLflow artifacts, make sure to restrict the scope of your server's cloud credentials such that it can only access files and directories related to MLflow.References
More information about the vulnerability is available at https://nvd.nist.gov/vuln/detail/CVE-2023-1177.
CVE-2023-2356
Relative Path Traversal in GitHub repository mlflow/mlflow prior to 2.3.1.
GHSA-83fm-w79m-64r5
Impact
Users of the MLflow Open Source Project who are hosting the MLflow Model Registry using the
mlflow serverormlflow uicommands using an MLflow version older than MLflow 2.3.1 may be vulnerable to a remote file access exploit if they are not limiting who can query their server (for example, by using a cloud VPC, an IP allowlist for inbound requests, or authentication / authorization middleware).This issue only affects users and integrations that run the
mlflow serverandmlflow uicommands. Integrations that do not make use ofmlflow serverormlflow uiare unaffected; for example, the Databricks Managed MLflow product and MLflow on Azure Machine Learning do not make use of these commands and are not impacted by these vulnerabilities in any way.The vulnerability is very similar to https://nvd.nist.gov/vuln/detail/CVE-2023-1177, and a separate CVE will be published and updated here shortly.
Patches
This vulnerability has been patched in MLflow 2.3.1, which was released to PyPI on April 27th, 2023. If you are using
mlflow serverormlflow uiwith the MLflow Model Registry, we recommend upgrading to MLflow 2.3.1 as soon as possible.Workarounds
If you are using the MLflow open source
mlflow serverormlflow uicommands, we strongly recommend limiting who can access your MLflow Model Registry and MLflow Tracking servers using a cloud VPC, an IP allowlist for inbound requests, authentication / authorization middleware, or another access restriction mechanism of your choosing.If you are using the MLflow open source
mlflow serverormlflow uicommands, we also strongly recommend limiting the remote files to which your MLflow Model Registry and MLflow Tracking servers have access. For example, if your MLflow Model Registry or MLflow Tracking server uses cloud-hosted blob storage for MLflow artifacts, make sure to restrict the scope of your server's cloud credentials such that it can only access files and directories related to MLflow.References
CVE-2023-2780
mlflow prior to 2.3.0 is vulnerable to path traversal due to a bypass of the fix for CVE-2023-1177.
CVE-2023-30172
A directory traversal vulnerability in the /get-artifact API method of the mlflow platform prior to v2.0.0 allows attackers to read arbitrary files on the server via the path parameter.
CVE-2023-3765
Absolute Path Traversal in GitHub repository mlflow/mlflow prior to 2.5.0.
Release Notes
mlflow/mlflow (mlflow)
### [`v2.5.0`](https://togithub.com/mlflow/mlflow/blob/HEAD/CHANGELOG.md#250-2023-07-17) [Compare Source](https://togithub.com/mlflow/mlflow/compare/v2.4.2...v2.5.0) MLflow 2.5.0 includes several major features and improvements: - \[MLflow AI Gateway] We are excited to announce the release of MLflow AI Gateway, a powerful tool designed to streamline the usage and management of various large language model (LLM) providers, such as OpenAI and Anthropic, within an organization. It offers a standardized interface that simplifies the interaction with these services and delivers centralized, secure management of credentials. To get started with MLflow AI Gateway, check out the docs at https://mlflow.org/docs/latest/gateway/index.html. ([#8694](https://togithub.com/mlflow/mlflow/issues/8694), [@harupy](https://togithub.com/harupy), [@BenWilson2](https://togithub.com/BenWilson2), [@dbczumar](https://togithub.com/dbczumar)) - \[Auth]: We are excited to announce the release of authentication and authorization support for MLflow Tracking and the MLflow Model Registry, providing integrated access control capabilities to both services. To get started, check out the docs at https://mlflow.org/docs/latest/auth/index.html. ([#9000](https://togithub.com/mlflow/mlflow/issues/9000), [#8975](https://togithub.com/mlflow/mlflow/issues/8975), [#8626](https://togithub.com/mlflow/mlflow/issues/8626), [#8837](https://togithub.com/mlflow/mlflow/issues/8837), [#8841](https://togithub.com/mlflow/mlflow/issues/8841), [@gabrielfu](https://togithub.com/gabrielfu), [@harupy](https://togithub.com/harupy)) Features: - \[Models] Add Support to the LangChain flavor for chains that contain unserializable components ([#8736](https://togithub.com/mlflow/mlflow/issues/8736), [@liangz1](https://togithub.com/liangz1)) - \[Scoring] Infer spark udf return type from model output schema ([#8934](https://togithub.com/mlflow/mlflow/issues/8934), [@WeichenXu123](https://togithub.com/WeichenXu123)) - \[Models] Add support for automated signature inference ([#8860](https://togithub.com/mlflow/mlflow/issues/8860), [#8782](https://togithub.com/mlflow/mlflow/issues/8782) [#8795](https://togithub.com/mlflow/mlflow/issues/8795), [#8725](https://togithub.com/mlflow/mlflow/issues/8725), [@jerrylian-db](https://togithub.com/jerrylian-db)) Bug fixes: - \[Security] Improve robustness to LFI attacks on Windows by enhancing path validation ([#8999](https://togithub.com/mlflow/mlflow/issues/8999), [@serena-ruan](https://togithub.com/serena-ruan)) - If you are using `mlflow server` or `mlflow ui` on Windows, we recommend upgrading to MLflow 2.5.0 as soon as possible. - \[Scoring] Support nullable array type values as spark_udf return values ([#9014](https://togithub.com/mlflow/mlflow/issues/9014), [@WeichenXu123](https://togithub.com/WeichenXu123)) - \[Models] Revert cache deletion of system modules when adding custom model code to the system path ([#8722](https://togithub.com/mlflow/mlflow/issues/8722), [@trungn1](https://togithub.com/trungn1)) - \[Models] add micro version to mlflow version pinning ([#8687](https://togithub.com/mlflow/mlflow/issues/8687), [@C-K-Loan](https://togithub.com/C-K-Loan)) - \[Artifacts] Prevent manually deleted artifacts from causing artifact garbage collection to fail ([#8498](https://togithub.com/mlflow/mlflow/issues/8498), [@PenHsuanWang](https://togithub.com/PenHsuanWang)) Documentation updates: - \[Docs] Update .push_model_to_sagemaker docs ([#8851](https://togithub.com/mlflow/mlflow/issues/8851), [@pdifranc](https://togithub.com/pdifranc)) - \[Docs] Fix invalid link for Azure ML documentation ([#8800](https://togithub.com/mlflow/mlflow/issues/8800), [@dunnkers](https://togithub.com/dunnkers)) - \[Artifacts / Docs / Models / Projects] Adds information on the OCI MLflow plugins for seamless integration with Oralce Cloud Infrastructure services. ([#8707](https://togithub.com/mlflow/mlflow/issues/8707), [@mrDzurb](https://togithub.com/mrDzurb)) Deprecation: - \[Models] Deprecate the `gluon` model flavor. The `mlflow.gluon` module will be removed in a future release. ([#8968](https://togithub.com/mlflow/mlflow/issues/8968), [@harupy](https://togithub.com/harupy)) Small bug fixes and documentation updates: [#9069](https://togithub.com/mlflow/mlflow/issues/9069), [#9056](https://togithub.com/mlflow/mlflow/issues/9056), [#9055](https://togithub.com/mlflow/mlflow/issues/9055), [#9054](https://togithub.com/mlflow/mlflow/issues/9054), [#9048](https://togithub.com/mlflow/mlflow/issues/9048), [#9043](https://togithub.com/mlflow/mlflow/issues/9043), [#9035](https://togithub.com/mlflow/mlflow/issues/9035), [#9034](https://togithub.com/mlflow/mlflow/issues/9034), [#9037](https://togithub.com/mlflow/mlflow/issues/9037), [#9038](https://togithub.com/mlflow/mlflow/issues/9038), [#8993](https://togithub.com/mlflow/mlflow/issues/8993), [#8966](https://togithub.com/mlflow/mlflow/issues/8966), [#8985](https://togithub.com/mlflow/mlflow/issues/8985), [@BenWilson2](https://togithub.com/BenWilson2); [#9039](https://togithub.com/mlflow/mlflow/issues/9039), [#9036](https://togithub.com/mlflow/mlflow/issues/9036), [#8902](https://togithub.com/mlflow/mlflow/issues/8902), [#8924](https://togithub.com/mlflow/mlflow/issues/8924), [#8866](https://togithub.com/mlflow/mlflow/issues/8866), [#8861](https://togithub.com/mlflow/mlflow/issues/8861), [#8810](https://togithub.com/mlflow/mlflow/issues/8810), [#8761](https://togithub.com/mlflow/mlflow/issues/8761), [#8544](https://togithub.com/mlflow/mlflow/issues/8544), [@jerrylian-db](https://togithub.com/jerrylian-db); [#8903](https://togithub.com/mlflow/mlflow/issues/8903), [@smurching](https://togithub.com/smurching); [#9080](https://togithub.com/mlflow/mlflow/issues/9080), [#9079](https://togithub.com/mlflow/mlflow/issues/9079), [#9078](https://togithub.com/mlflow/mlflow/issues/9078), [#9076](https://togithub.com/mlflow/mlflow/issues/9076), [#9075](https://togithub.com/mlflow/mlflow/issues/9075), [#9074](https://togithub.com/mlflow/mlflow/issues/9074), [#9071](https://togithub.com/mlflow/mlflow/issues/9071), [#9063](https://togithub.com/mlflow/mlflow/issues/9063), [#9062](https://togithub.com/mlflow/mlflow/issues/9062), [#9032](https://togithub.com/mlflow/mlflow/issues/9032), [#9031](https://togithub.com/mlflow/mlflow/issues/9031), [#9027](https://togithub.com/mlflow/mlflow/issues/9027), [#9023](https://togithub.com/mlflow/mlflow/issues/9023), [#9022](https://togithub.com/mlflow/mlflow/issues/9022), [#9020](https://togithub.com/mlflow/mlflow/issues/9020), [#9005](https://togithub.com/mlflow/mlflow/issues/9005), [#8994](https://togithub.com/mlflow/mlflow/issues/8994), [#8979](https://togithub.com/mlflow/mlflow/issues/8979), [#8983](https://togithub.com/mlflow/mlflow/issues/8983), [#8984](https://togithub.com/mlflow/mlflow/issues/8984), [#8982](https://togithub.com/mlflow/mlflow/issues/8982), [#8970](https://togithub.com/mlflow/mlflow/issues/8970), [#8962](https://togithub.com/mlflow/mlflow/issues/8962), [#8969](https://togithub.com/mlflow/mlflow/issues/8969), [#8968](https://togithub.com/mlflow/mlflow/issues/8968), [#8959](https://togithub.com/mlflow/mlflow/issues/8959), [#8960](https://togithub.com/mlflow/mlflow/issues/8960), [#8958](https://togithub.com/mlflow/mlflow/issues/8958), [#8956](https://togithub.com/mlflow/mlflow/issues/8956), [#8955](https://togithub.com/mlflow/mlflow/issues/8955), [#8954](https://togithub.com/mlflow/mlflow/issues/8954), [#8949](https://togithub.com/mlflow/mlflow/issues/8949), [#8950](https://togithub.com/mlflow/mlflow/issues/8950), [#8952](https://togithub.com/mlflow/mlflow/issues/8952), [#8948](https://togithub.com/mlflow/mlflow/issues/8948), [#8946](https://togithub.com/mlflow/mlflow/issues/8946), [#8947](https://togithub.com/mlflow/mlflow/issues/8947), [#8943](https://togithub.com/mlflow/mlflow/issues/8943), [#8944](https://togithub.com/mlflow/mlflow/issues/8944), [#8916](https://togithub.com/mlflow/mlflow/issues/8916), [#8917](https://togithub.com/mlflow/mlflow/issues/8917), [#8933](https://togithub.com/mlflow/mlflow/issues/8933), [#8929](https://togithub.com/mlflow/mlflow/issues/8929), [#8932](https://togithub.com/mlflow/mlflow/issues/8932), [#8927](https://togithub.com/mlflow/mlflow/issues/8927), [#8930](https://togithub.com/mlflow/mlflow/issues/8930), [#8925](https://togithub.com/mlflow/mlflow/issues/8925), [#8921](https://togithub.com/mlflow/mlflow/issues/8921), [#8873](https://togithub.com/mlflow/mlflow/issues/8873), [#8915](https://togithub.com/mlflow/mlflow/issues/8915), [#8909](https://togithub.com/mlflow/mlflow/issues/8909), [#8908](https://togithub.com/mlflow/mlflow/issues/8908), [#8911](https://togithub.com/mlflow/mlflow/issues/8911), [#8910](https://togithub.com/mlflow/mlflow/issues/8910), [#8907](https://togithub.com/mlflow/mlflow/issues/8907), [#8906](https://togithub.com/mlflow/mlflow/issues/8906), [#8898](https://togithub.com/mlflow/mlflow/issues/8898), [#8893](https://togithub.com/mlflow/mlflow/issues/8893), [#8889](https://togithub.com/mlflow/mlflow/issues/8889), [#8892](https://togithub.com/mlflow/mlflow/issues/8892), [#8891](https://togithub.com/mlflow/mlflow/issues/8891), [#8887](https://togithub.com/mlflow/mlflow/issues/8887), [#8875](https://togithub.com/mlflow/mlflow/issues/8875), [#8876](https://togithub.com/mlflow/mlflow/issues/8876), [#8882](https://togithub.com/mlflow/mlflow/issues/8882), [#8874](https://togithub.com/mlflow/mlflow/issues/8874), [#8868](https://togithub.com/mlflow/mlflow/issues/8868), [#8872](https://togithub.com/mlflow/mlflow/issues/8872), [#8869](https://togithub.com/mlflow/mlflow/issues/8869), [#8828](https://togithub.com/mlflow/mlflow/issues/8828), [#8852](https://togithub.com/mlflow/mlflow/issues/8852), [#8857](https://togithub.com/mlflow/mlflow/issues/8857), [#8853](https://togithub.com/mlflow/mlflow/issues/8853), [#8854](https://togithub.com/mlflow/mlflow/issues/8854), [#8848](https://togithub.com/mlflow/mlflow/issues/8848), [#8850](https://togithub.com/mlflow/mlflow/issues/8850), [#8840](https://togithub.com/mlflow/mlflow/issues/8840), [#8835](https://togithub.com/mlflow/mlflow/issues/8835), [#8832](https://togithub.com/mlflow/mlflow/issues/8832), [#8831](https://togithub.com/mlflow/mlflow/issues/8831), [#8830](https://togithub.com/mlflow/mlflow/issues/8830), [#8829](https://togithub.com/mlflow/mlflow/issues/8829), [#8839](https://togithub.com/mlflow/mlflow/issues/8839), [#8833](https://togithub.com/mlflow/mlflow/issues/8833), [#8838](https://togithub.com/mlflow/mlflow/issues/8838), [#8819](https://togithub.com/mlflow/mlflow/issues/8819), [#8814](https://togithub.com/mlflow/mlflow/issues/8814), [#8825](https://togithub.com/mlflow/mlflow/issues/8825), [#8818](https://togithub.com/mlflow/mlflow/issues/8818), [#8787](https://togithub.com/mlflow/mlflow/issues/8787), [#8775](https://togithub.com/mlflow/mlflow/issues/8775), [#8749](https://togithub.com/mlflow/mlflow/issues/8749), [#8766](https://togithub.com/mlflow/mlflow/issues/8766), [#8756](https://togithub.com/mlflow/mlflow/issues/8756), [#8753](https://togithub.com/mlflow/mlflow/issues/8753), [#8751](https://togithub.com/mlflow/mlflow/issues/8751), [#8748](https://togithub.com/mlflow/mlflow/issues/8748), [#8744](https://togithub.com/mlflow/mlflow/issues/8744), [#8731](https://togithub.com/mlflow/mlflow/issues/8731), [#8717](https://togithub.com/mlflow/mlflow/issues/8717), [#8730](https://togithub.com/mlflow/mlflow/issues/8730), [#8691](https://togithub.com/mlflow/mlflow/issues/8691), [#8720](https://togithub.com/mlflow/mlflow/issues/8720), [#8723](https://togithub.com/mlflow/mlflow/issues/8723), [#8719](https://togithub.com/mlflow/mlflow/issues/8719), [#8688](https://togithub.com/mlflow/mlflow/issues/8688), [#8721](https://togithub.com/mlflow/mlflow/issues/8721), [#8715](https://togithub.com/mlflow/mlflow/issues/8715), [#8716](https://togithub.com/mlflow/mlflow/issues/8716), [#8718](https://togithub.com/mlflow/mlflow/issues/8718), [#8696](https://togithub.com/mlflow/mlflow/issues/8696), [#8698](https://togithub.com/mlflow/mlflow/issues/8698), [#8692](https://togithub.com/mlflow/mlflow/issues/8692), [#8693](https://togithub.com/mlflow/mlflow/issues/8693), [#8690](https://togithub.com/mlflow/mlflow/issues/8690), [@harupy](https://togithub.com/harupy); [#9030](https://togithub.com/mlflow/mlflow/issues/9030), [@AlimurtuzaCodes](https://togithub.com/AlimurtuzaCodes); [#9029](https://togithub.com/mlflow/mlflow/issues/9029), [#9025](https://togithub.com/mlflow/mlflow/issues/9025), [#9021](https://togithub.com/mlflow/mlflow/issues/9021), [#9013](https://togithub.com/mlflow/mlflow/issues/9013), [@viktoriussuwandi](https://togithub.com/viktoriussuwandi); [#9010](https://togithub.com/mlflow/mlflow/issues/9010), [@Bncer](https://togithub.com/Bncer); [#9011](https://togithub.com/mlflow/mlflow/issues/9011), [@Pecunia201](https://togithub.com/Pecunia201); [#9007](https://togithub.com/mlflow/mlflow/issues/9007), [#9003](https://togithub.com/mlflow/mlflow/issues/9003), [@EdAbati](https://togithub.com/EdAbati); [#9002](https://togithub.com/mlflow/mlflow/issues/9002), [@prithvikannan](https://togithub.com/prithvikannan); [#8991](https://togithub.com/mlflow/mlflow/issues/8991), [#8867](https://togithub.com/mlflow/mlflow/issues/8867), [@AveshCSingh](https://togithub.com/AveshCSingh); [#8951](https://togithub.com/mlflow/mlflow/issues/8951), [#8896](https://togithub.com/mlflow/mlflow/issues/8896), [#8888](https://togithub.com/mlflow/mlflow/issues/8888), [#8849](https://togithub.com/mlflow/mlflow/issues/8849), [@gabrielfu](https://togithub.com/gabrielfu); [#8913](https://togithub.com/mlflow/mlflow/issues/8913), [#8885](https://togithub.com/mlflow/mlflow/issues/8885), [#8871](https://togithub.com/mlflow/mlflow/issues/8871), [#8870](https://togithub.com/mlflow/mlflow/issues/8870), [#8788](https://togithub.com/mlflow/mlflow/issues/8788), [#8772](https://togithub.com/mlflow/mlflow/issues/8772), [#8771](https://togithub.com/mlflow/mlflow/issues/8771), [@serena-ruan](https://togithub.com/serena-ruan); [#8879](https://togithub.com/mlflow/mlflow/issues/8879), [@maciejskorski](https://togithub.com/maciejskorski); [#7752](https://togithub.com/mlflow/mlflow/issues/7752), [@arunkumarkota](https://togithub.com/arunkumarkota); [#9083](https://togithub.com/mlflow/mlflow/issues/9083), [#9081](https://togithub.com/mlflow/mlflow/issues/9081), [#8765](https://togithub.com/mlflow/mlflow/issues/8765), [#8742](https://togithub.com/mlflow/mlflow/issues/8742), [#8685](https://togithub.com/mlflow/mlflow/issues/8685), [#8682](https://togithub.com/mlflow/mlflow/issues/8682), [#8683](https://togithub.com/mlflow/mlflow/issues/8683), [@dbczumar](https://togithub.com/dbczumar); [#8791](https://togithub.com/mlflow/mlflow/issues/8791), [@mhattingpete](https://togithub.com/mhattingpete); [#8739](https://togithub.com/mlflow/mlflow/issues/8739), [@yunpark93](https://togithub.com/yunpark93) ### [`v2.4.2`](https://togithub.com/mlflow/mlflow/blob/HEAD/CHANGELOG.md#242-2023-07-10) [Compare Source](https://togithub.com/mlflow/mlflow/compare/v2.4.1...v2.4.2) MLflow 2.4.2 is a patch release containing the following bug fixes and changes: Bug fixes: - \[Models] Add compatibility for legacy transformers serialization ([#8964](https://togithub.com/mlflow/mlflow/issues/8964), [@BenWilson2](https://togithub.com/BenWilson2)) - \[Models] Fix downloading MLmodel files from alias-based models:/ URIs ([#8764](https://togithub.com/mlflow/mlflow/issues/8764), [@smurching](https://togithub.com/smurching)) - \[Models] Fix reading model flavor config from URI for models in UC ([#8728](https://togithub.com/mlflow/mlflow/issues/8728), [@smurching](https://togithub.com/smurching)) - \[Models] Support `feature_deps` in ModelVersion creation for UC ([#8867](https://togithub.com/mlflow/mlflow/issues/8867), [#8815](https://togithub.com/mlflow/mlflow/issues/8815), [@AveshCSingh](https://togithub.com/AveshCSingh)) - \[Models] Add support for listing artifacts in UC model registry artifact repo ([#8803](https://togithub.com/mlflow/mlflow/issues/8803), [@smurching](https://togithub.com/smurching)) - \[Core] Include resources for recipes in mlflow-skinny ([#8895](https://togithub.com/mlflow/mlflow/issues/8895), [@harupy](https://togithub.com/harupy)) - \[UI] Enable datasets tracking UI ([#8886](https://togithub.com/mlflow/mlflow/issues/8886), [@harupy](https://togithub.com/harupy)) - \[Artifacts] Use `MLFLOW_ENABLE_MULTIPART_DOWNLOAD` in `DatabricksArtifactRepository` ([#8884](https://togithub.com/mlflow/mlflow/issues/8884), [@harupy](https://togithub.com/harupy)) Documentation updates: - \[Examples / Docs] Add question-answering and summarization examples and docs with LLMs ([#8695](https://togithub.com/mlflow/mlflow/issues/8695), [@dbczumar](https://togithub.com/dbczumar)) - \[Examples / Docs] Add johnsnowlabs flavor example and doc ([#8689](https://togithub.com/mlflow/mlflow/issues/8689), [@C-K-Loan](https://togithub.com/C-K-Loan)) Small bug fixes and documentation updates: [#8966](https://togithub.com/mlflow/mlflow/issues/8966), [@BenWilson2](https://togithub.com/BenWilson2); [#8881](https://togithub.com/mlflow/mlflow/issues/8881), [@harupy](https://togithub.com/harupy); [#8846](https://togithub.com/mlflow/mlflow/issues/8846), [#8760](https://togithub.com/mlflow/mlflow/issues/8760), [@smurching](https://togithub.com/smurching) ### [`v2.4.1`](https://togithub.com/mlflow/mlflow/blob/HEAD/CHANGELOG.md#241-2023-06-09) [Compare Source](https://togithub.com/mlflow/mlflow/compare/v2.4.0...v2.4.1) MLflow 2.4.1 is a patch release containing the following features, bug fixes and changes: Features: - \[Tracking] Extend SearchRuns to support datasets ([#8622](https://togithub.com/mlflow/mlflow/issues/8622), [@prithvikannan](https://togithub.com/prithvikannan)) - \[Models] Add an `mlflow.johnsnowlabs` flavor for the `johnsnowlabs` package ([#8556](https://togithub.com/mlflow/mlflow/issues/8556), [@C-K-Loan](https://togithub.com/C-K-Loan)) - \[Models] Add a warning for duplicate pip requirements specified in `save_model` and `log_model` for the `transformers` flavor ([#8678](https://togithub.com/mlflow/mlflow/issues/8678), [@BenWilson2](https://togithub.com/BenWilson2)) Bug fixes: - \[Security] Improve robustness to LFI attacks ([#8648](https://togithub.com/mlflow/mlflow/issues/8648), [@serena-ruan](https://togithub.com/serena-ruan)) - If you are using `mlflow server` or `mlflow ui`, we recommend upgrading to MLflow 2.4.1 as soon as possible. - \[Models] Fix an issue with `transformers` serialization for ModelCards that contain invalid characters ([#8652](https://togithub.com/mlflow/mlflow/issues/8652), [@BenWilson2](https://togithub.com/BenWilson2)) - \[Models] Fix connection pooling deadlocks that occurred during large file downloads ([#8682](https://togithub.com/mlflow/mlflow/issues/8682), [@dbczumar](https://togithub.com/dbczumar); [#8660](https://togithub.com/mlflow/mlflow/issues/8660), [@harupy](https://togithub.com/harupy)) Small bug fixes and documentation updates: [#8677](https://togithub.com/mlflow/mlflow/issues/8677), [#8674](https://togithub.com/mlflow/mlflow/issues/8674), [#8646](https://togithub.com/mlflow/mlflow/issues/8646), [#8647](https://togithub.com/mlflow/mlflow/issues/8647), [@dbczumar](https://togithub.com/dbczumar); [#8654](https://togithub.com/mlflow/mlflow/issues/8654), [#8653](https://togithub.com/mlflow/mlflow/issues/8653), [#8660](https://togithub.com/mlflow/mlflow/issues/8660), [#8650](https://togithub.com/mlflow/mlflow/issues/8650), [#8642](https://togithub.com/mlflow/mlflow/issues/8642), [#8636](https://togithub.com/mlflow/mlflow/issues/8636), [#8599](https://togithub.com/mlflow/mlflow/issues/8599), [#8637](https://togithub.com/mlflow/mlflow/issues/8637), [#8608](https://togithub.com/mlflow/mlflow/issues/8608), [#8633](https://togithub.com/mlflow/mlflow/issues/8633), [#8623](https://togithub.com/mlflow/mlflow/issues/8623), [#8628](https://togithub.com/mlflow/mlflow/issues/8628), [#8619](https://togithub.com/mlflow/mlflow/issues/8619), [@harupy](https://togithub.com/harupy); [#8655](https://togithub.com/mlflow/mlflow/issues/8655), [#8609](https://togithub.com/mlflow/mlflow/issues/8609), [@BenWilson2](https://togithub.com/BenWilson2); [#8648](https://togithub.com/mlflow/mlflow/issues/8648), [@serena-ruan](https://togithub.com/serena-ruan); [#8521](https://togithub.com/mlflow/mlflow/issues/8521), [@ka1mar](https://togithub.com/ka1mar); [#8638](https://togithub.com/mlflow/mlflow/issues/8638), [@smurching](https://togithub.com/smurching); [#8634](https://togithub.com/mlflow/mlflow/issues/8634), [@PenHsuanWang](https://togithub.com/PenHsuanWang) ### [`v2.4.0`](https://togithub.com/mlflow/mlflow/blob/HEAD/CHANGELOG.md#240-2023-06-06) [Compare Source](https://togithub.com/mlflow/mlflow/compare/v2.3.2...v2.4.0) MLflow 2.4.0 includes several major features and improvements Features: - \[Tracking] Introduce dataset tracking APIs: `mlflow.data` and `mlflow.log_input()` ([#8186](https://togithub.com/mlflow/mlflow/issues/8186), [@prithvikannan](https://togithub.com/prithvikannan)) - \[Tracking] Add `mlflow.log_table()` and `mlflow.load_table()` APIs for logging evaluation tables ([#8523](https://togithub.com/mlflow/mlflow/issues/8523), [#8467](https://togithub.com/mlflow/mlflow/issues/8467), [@sunishsheth2009](https://togithub.com/sunishsheth2009)) - \[Tracking] Introduce `mlflow.get_parent_run()` fluent API ([#8493](https://togithub.com/mlflow/mlflow/issues/8493), [@annzhang-db](https://togithub.com/annzhang-db)) - \[Tracking / Model Registry] Re-introduce faster artifact downloads on Databricks ([#8352](https://togithub.com/mlflow/mlflow/issues/8352), [@dbczumar](https://togithub.com/dbczumar); [#8561](https://togithub.com/mlflow/mlflow/issues/8561), [@harupy](https://togithub.com/harupy)) - \[UI] Add dataset tracking information to MLflow Tracking UI ([#8602](https://togithub.com/mlflow/mlflow/issues/8602), [@prithvikannan](https://togithub.com/prithvikannan), [@hubertzub-db](https://togithub.com/hubertzub-db)) - \[UI] Introduce Artifact View for comparing inputs, outputs, and metadata across models ([#8602](https://togithub.com/mlflow/mlflow/issues/8602), [@hubertzub-db](https://togithub.com/hubertzub-db)) - \[Models] Extend `mlflow.evaluate()` to support LLM tasks ([#8484](https://togithub.com/mlflow/mlflow/issues/8484), [@harupy](https://togithub.com/harupy)) - \[Models] Support logging subclasses of `Chain` and `LLMChain` in `mlflow.langchain` flavor ([#8453](https://togithub.com/mlflow/mlflow/issues/8453), [@liangz1](https://togithub.com/liangz1)) - \[Models] Add support for LangChain Agents to the `mlflow.langchain` flavor ([#8297](https://togithub.com/mlflow/mlflow/issues/8297), [@sunishsheth2009](https://togithub.com/sunishsheth2009)) - \[Models] Add a `mlflow.sentence_transformers` flavor for SentenceTransformers ([#8479](https://togithub.com/mlflow/mlflow/issues/8479), [@BenWilson2](https://togithub.com/BenWilson2); [#8547](https://togithub.com/mlflow/mlflow/issues/8547), [@Loquats](https://togithub.com/Loquats)) - \[Models] Add support for multi-GPU inference and efficient weight loading for `mlflow.transformers` flavor ([#8448](https://togithub.com/mlflow/mlflow/issues/8448), [@ankit-db](https://togithub.com/ankit-db)) - \[Models] Support the `max_shard_size` parameter in the `mlflow.transformers` flavor ([#8567](https://togithub.com/mlflow/mlflow/issues/8567), [@wenfeiy-db](https://togithub.com/wenfeiy-db)) - \[Models] Add support for audio transcription pipelines in the `mlflow.transformers` flavor ([#8464](https://togithub.com/mlflow/mlflow/issues/8464), [@BenWilson2](https://togithub.com/BenWilson2)) - \[Models] Add support for audio classification to `mlflow.transformers` flavor ([#8492](https://togithub.com/mlflow/mlflow/issues/8492), [@BenWilson2](https://togithub.com/BenWilson2)) - \[Models] Add support for URI inputs in audio models logged with the `mlflow.transformers` flavor ([#8495](https://togithub.com/mlflow/mlflow/issues/8495), [@BenWilson2](https://togithub.com/BenWilson2)) - \[Models] Add support for returning classifier scores in `mlflow.transformers` pyfunc outputs ([#8512](https://togithub.com/mlflow/mlflow/issues/8512), [@BenWilson2](https://togithub.com/BenWilson2)) - \[Models] Support optional inputs in model signatures ([#8438](https://togithub.com/mlflow/mlflow/issues/8438), [@apurva-koti](https://togithub.com/apurva-koti)) - \[Models] Introduce an `mlflow.models.set_signature()` API to set the signature of a logged model ([#8476](https://togithub.com/mlflow/mlflow/issues/8476), [@jerrylian-db](https://togithub.com/jerrylian-db)) - \[Models] Persist ONNX Runtime InferenceSession options when logging a model with `mlflow.onnx.log_model()` ([#8433](https://togithub.com/mlflow/mlflow/issues/8433), [@leqiao-1](https://togithub.com/leqiao-1)) Bug fixes: - \[Tracking] Terminate Spark callback server when Spark Autologging is disabled or Spark Session is shut down ([#8508](https://togithub.com/mlflow/mlflow/issues/8508), [@WeichenXu123](https://togithub.com/WeichenXu123)) - \[Tracking] Fix compatibility of `mlflow server` with `Flask<2.0` ([#8463](https://togithub.com/mlflow/mlflow/issues/8463), [@kevingreer](https://togithub.com/kevingreer)) - \[Models] Convert `mlflow.transformers` pyfunc scalar string output to list of strings during batch inference ([#8546](https://togithub.com/mlflow/mlflow/issues/8546), [@BenWilson2](https://togithub.com/BenWilson2)) - \[Models] Fix a bug causing outdated pyenv versions to be installed by `mlflow models build-docker` ([#8488](https://togithub.com/mlflow/mlflow/issues/8488), [@Hellzed](https://togithub.com/Hellzed)) - \[Model Registry] Remove aliases from storage when a Model Version is deleted ([#8459](https://togithub.com/mlflow/mlflow/issues/8459), [@arpitjasa-db](https://togithub.com/arpitjasa-db)) Documentation updates: - \[Docs] Publish a new MLOps Quickstart for model selection and deployment ([#8462](https://togithub.com/mlflow/mlflow/issues/8462), [@lobrien](https://togithub.com/lobrien)) - \[Docs] Add MLflavors library to Community Model Flavors documentation ([#8420](https://togithub.com/mlflow/mlflow/issues/8420), [@benjaminbluhm](https://togithub.com/benjaminbluhm)) - \[Docs] Add documentation for Registered Model Aliases ([#8445](https://togithub.com/mlflow/mlflow/issues/8445), [@arpitjasa-db](https://togithub.com/arpitjasa-db)) - \[Docs] Fix errors in documented `mlflow models` CLI command examples ([#8480](https://togithub.com/mlflow/mlflow/issues/8480), [@vijethmoudgalya](https://togithub.com/vijethmoudgalya)) Small bug fixes and documentation updates: [#8611](https://togithub.com/mlflow/mlflow/issues/8611), [#8587](https://togithub.com/mlflow/mlflow/issues/8587), [@dbczumar](https://togithub.com/dbczumar); [#8617](https://togithub.com/mlflow/mlflow/issues/8617), [#8620](https://togithub.com/mlflow/mlflow/issues/8620), [#8615](https://togithub.com/mlflow/mlflow/issues/8615), [#8603](https://togithub.com/mlflow/mlflow/issues/8603), [#8604](https://togithub.com/mlflow/mlflow/issues/8604), [#8601](https://togithub.com/mlflow/mlflow/issues/8601), [#8596](https://togithub.com/mlflow/mlflow/issues/8596), [#8598](https://togithub.com/mlflow/mlflow/issues/8598), [#8597](https://togithub.com/mlflow/mlflow/issues/8597), [#8589](https://togithub.com/mlflow/mlflow/issues/8589), [#8580](https://togithub.com/mlflow/mlflow/issues/8580), [#8581](https://togithub.com/mlflow/mlflow/issues/8581), [#8575](https://togithub.com/mlflow/mlflow/issues/8575), [#8582](https://togithub.com/mlflow/mlflow/issues/8582), [#8577](https://togithub.com/mlflow/mlflow/issues/8577), [#8576](https://togithub.com/mlflow/mlflow/issues/8576), [#8578](https://togithub.com/mlflow/mlflow/issues/8578), [#8561](https://togithub.com/mlflow/mlflow/issues/8561), [#8568](https://togithub.com/mlflow/mlflow/issues/8568), [#8551](https://togithub.com/mlflow/mlflow/issues/8551), [#8528](https://togithub.com/mlflow/mlflow/issues/8528), [#8550](https://togithub.com/mlflow/mlflow/issues/8550), [#8489](https://togithub.com/mlflow/mlflow/issues/8489), [#8530](https://togithub.com/mlflow/mlflow/issues/8530), [#8534](https://togithub.com/mlflow/mlflow/issues/8534), [#8533](https://togithub.com/mlflow/mlflow/issues/8533), [#8532](https://togithub.com/mlflow/mlflow/issues/8532), [#8524](https://togithub.com/mlflow/mlflow/issues/8524), [#8520](https://togithub.com/mlflow/mlflow/issues/8520), [#8517](https://togithub.com/mlflow/mlflow/issues/8517), [#8516](https://togithub.com/mlflow/mlflow/issues/8516), [#8515](https://togithub.com/mlflow/mlflow/issues/8515), [#8514](https://togithub.com/mlflow/mlflow/issues/8514), [#8506](https://togithub.com/mlflow/mlflow/issues/8506), [#8503](https://togithub.com/mlflow/mlflow/issues/8503), [#8500](https://togithub.com/mlflow/mlflow/issues/8500), [#8504](https://togithub.com/mlflow/mlflow/issues/8504), [#8496](https://togithub.com/mlflow/mlflow/issues/8496), [#8486](https://togithub.com/mlflow/mlflow/issues/8486), [#8485](https://togithub.com/mlflow/mlflow/issues/8485), [#8468](https://togithub.com/mlflow/mlflow/issues/8468), [#8471](https://togithub.com/mlflow/mlflow/issues/8471), [#8473](https://togithub.com/mlflow/mlflow/issues/8473), [#8470](https://togithub.com/mlflow/mlflow/issues/8470), [#8458](https://togithub.com/mlflow/mlflow/issues/8458), [#8447](https://togithub.com/mlflow/mlflow/issues/8447), [#8446](https://togithub.com/mlflow/mlflow/issues/8446), [#8434](https://togithub.com/mlflow/mlflow/issues/8434), [@harupy](https://togithub.com/harupy); [#8607](https://togithub.com/mlflow/mlflow/issues/8607), [#8538](https://togithub.com/mlflow/mlflow/issues/8538), [#8513](https://togithub.com/mlflow/mlflow/issues/8513), [#8452](https://togithub.com/mlflow/mlflow/issues/8452), [#8466](https://togithub.com/mlflow/mlflow/issues/8466), [#8465](https://togithub.com/mlflow/mlflow/issues/8465), [@serena-ruan](https://togithub.com/serena-ruan); [#8586](https://togithub.com/mlflow/mlflow/issues/8586), [#8595](https://togithub.com/mlflow/mlflow/issues/8595), [@prithvikannan](https://togithub.com/prithvikannan); [#8593](https://togithub.com/mlflow/mlflow/issues/8593), [#8541](https://togithub.com/mlflow/mlflow/issues/8541), [@kriscon-db](https://togithub.com/kriscon-db); [#8592](https://togithub.com/mlflow/mlflow/issues/8592), [#8566](https://togithub.com/mlflow/mlflow/issues/8566), [@annzhang-db](https://togithub.com/annzhang-db); [#8588](https://togithub.com/mlflow/mlflow/issues/8588), [#8565](https://togithub.com/mlflow/mlflow/issues/8565), [#8559](https://togithub.com/mlflow/mlflow/issues/8559), [#8537](https://togithub.com/mlflow/mlflow/issues/8537), [@BenWilson2](https://togithub.com/BenWilson2); [#8545](https://togithub.com/mlflow/mlflow/issues/8545), [@apurva-koti](https://togithub.com/apurva-koti); [#8564](https://togithub.com/mlflow/mlflow/issues/8564), [@DavidSpek](https://togithub.com/DavidSpek); [#8436](https://togithub.com/mlflow/mlflow/issues/8436), [#8490](https://togithub.com/mlflow/mlflow/issues/8490), [@jerrylian-db](https://togithub.com/jerrylian-db); [#8505](https://togithub.com/mlflow/mlflow/issues/8505), [@eliaskoromilas](https://togithub.com/eliaskoromilas); [#8483](https://togithub.com/mlflow/mlflow/issues/8483), [@WeichenXu123](https://togithub.com/WeichenXu123); [#8472](https://togithub.com/mlflow/mlflow/issues/8472), [@leqiao-1](https://togithub.com/leqiao-1); [#8429](https://togithub.com/mlflow/mlflow/issues/8429), [@jinzhang21](https://togithub.com/jinzhang21); [#8581](https://togithub.com/mlflow/mlflow/issues/8581), [#8548](https://togithub.com/mlflow/mlflow/issues/8548), [#8499](https://togithub.com/mlflow/mlflow/issues/8499), [@gabrielfu](https://togithub.com/gabrielfu); ### [`v2.3.2`](https://togithub.com/mlflow/mlflow/blob/HEAD/CHANGELOG.md#232-2023-05-12) [Compare Source](https://togithub.com/mlflow/mlflow/compare/v2.3.1...v2.3.2) MLflow 2.3.2 is a patch release containing the following features, bug fixes and changes: Features: - \[Models] Add GPU support for `transformers` models `pyfunc` inference and serving ([#8375](https://togithub.com/mlflow/mlflow/issues/8375), [@ankit-db](https://togithub.com/ankit-db)) - \[Models] Disable autologging functionality for non-relevant models when training a `transformers` model ([#8405](https://togithub.com/mlflow/mlflow/issues/8405), [@BenWilson2](https://togithub.com/BenWilson2)) - \[Models] Add support for preserving and overriding `torch_dtype` values in `transformers` pipelines ([#8421](https://togithub.com/mlflow/mlflow/issues/8421), [@BenWilson2](https://togithub.com/BenWilson2)) - \[Models] Add support for `Feature Extraction` pipelines in the `transformers` flavor ([#8423](https://togithub.com/mlflow/mlflow/issues/8423), [@BenWilson2](https://togithub.com/BenWilson2)) - \[Tracking] Add basic HTTP auth support for users, registered models, and experiments permissions ([#8286](https://togithub.com/mlflow/mlflow/issues/8286), [@gabrielfu](https://togithub.com/gabrielfu)) Bug Fixes: - \[Models] Fix inferred schema issue with `Text2TextGeneration` pipelines in the `transformers` flavor ([#8391](https://togithub.com/mlflow/mlflow/issues/8391), [@BenWilson2](https://togithub.com/BenWilson2)) - \[Models] Change MLflow dependency pinning in logged models from a range value to an exact major and minor version ([#8422](https://togithub.com/mlflow/mlflow/issues/8422), [@harupy](https://togithub.com/harupy)) Documentation updates: - \[Examples] Add `signature` logging to all examples and documentation ([#8410](https://togithub.com/mlflow/mlflow/issues/8410), [#8401](https://togithub.com/mlflow/mlflow/issues/8401), [#8400](https://togithub.com/mlflow/mlflow/issues/8400), [#8387](https://togithub.com/mlflow/mlflow/issues/8387) [@jerrylian-db](https://togithub.com/jerrylian-db)) - \[Examples] Add `sentence-transformers` examples to the `transformers` examples suite ([#8425](https://togithub.com/mlflow/mlflow/issues/8425), [@BenWilson2](https://togithub.com/BenWilson2)) - \[Docs] Add a new MLflow Quickstart documentation page ([#8171](https://togithub.com/mlflow/mlflow/issues/8171), [@lobrien](https://togithub.com/lobrien)) - \[Docs] Add a new introduction to MLflow page ([#8365](https://togithub.com/mlflow/mlflow/issues/8365), [@lobrien](https://togithub.com/lobrien)) - \[Docs] Add a community model plugin example and documentation for `trubrics` ([#8371](https://togithub.com/mlflow/mlflow/issues/8371), [@jeffkayne](https://togithub.com/jeffkayne)) - \[Docs] Add `gluon` pyfunc example to Model flavor documentation ([#8403](https://togithub.com/mlflow/mlflow/issues/8403), [@ericvincent18](https://togithub.com/ericvincent18)) - \[Docs] Add `statsmodels` pyfunc example to `Models` flavor documentation ([#8394](https://togithub.com/mlflow/mlflow/issues/8394), [@ericvincent18](https://togithub.com/ericvincent18)) Small bug fixes and documentation updates: [#8415](https://togithub.com/mlflow/mlflow/issues/8415), [#8412](https://togithub.com/mlflow/mlflow/issues/8412), [#8411](https://togithub.com/mlflow/mlflow/issues/8411), [#8355](https://togithub.com/mlflow/mlflow/issues/8355), [#8354](https://togithub.com/mlflow/mlflow/issues/8354), [#8353](https://togithub.com/mlflow/mlflow/issues/8353), [#8348](https://togithub.com/mlflow/mlflow/issues/8348), [@harupy](https://togithub.com/harupy); [#8374](https://togithub.com/mlflow/mlflow/issues/8374), [#8367](https://togithub.com/mlflow/mlflow/issues/8367), [#8350](https://togithub.com/mlflow/mlflow/issues/8350), [@dbczumar](https://togithub.com/dbczumar); [#8358](https://togithub.com/mlflow/mlflow/issues/8358) [@mrkaye97](https://togithub.com/mrkaye97); [#8392](https://togithub.com/mlflow/mlflow/issues/8392), [#8362](https://togithub.com/mlflow/mlflow/issues/8362), [@smurching](https://togithub.com/smurching); [#8427](https://togithub.com/mlflow/mlflow/issues/8427), [#8408](https://togithub.com/mlflow/mlflow/issues/8408), [#8399](https://togithub.com/mlflow/mlflow/issues/8399), [#8381](https://togithub.com/mlflow/mlflow/issues/8381), [@BenWilson2](https://togithub.com/BenWilson2); [#8395](https://togithub.com/mlflow/mlflow/issues/8395), [#8390](https://togithub.com/mlflow/mlflow/issues/8390), [@jerrylian-db](https://togithub.com/jerrylian-db); [#8402](https://togithub.com/mlflow/mlflow/issues/8402), [#8398](https://togithub.com/mlflow/mlflow/issues/8398), [@WeichenXu123](https://togithub.com/WeichenXu123); [#8377](https://togithub.com/mlflow/mlflow/issues/8377), [#8363](https://togithub.com/mlflow/mlflow/issues/8363), [@arpitjasa-db](https://togithub.com/arpitjasa-db); [#8385](https://togithub.com/mlflow/mlflow/issues/8385), [@prithvikannan](https://togithub.com/prithvikannan); [#8418](https://togithub.com/mlflow/mlflow/issues/8418), [@Jeukoh](https://togithub.com/Jeukoh); ### [`v2.3.1`](https://togithub.com/mlflow/mlflow/blob/HEAD/CHANGELOG.md#231-2023-04-27) [Compare Source](https://togithub.com/mlflow/mlflow/compare/v2.3.0...v2.3.1) MLflow 2.3.1 is a patch release containing the following bug fixes and changes: Bug fixes: - \[Security] Fix critical LFI attack vulnerability by disabling the ability to provide relative paths in registered model sources ([#8281](https://togithub.com/mlflow/mlflow/issues/8281), [@BenWilson2](https://togithub.com/BenWilson2)) - **If you are using `mlflow server` or `mlflow ui`, we recommend upgrading to MLflow 2.3.1 as soon as possible.** For more details, see https://github.com/mlflow/mlflow/security/advisories/GHSA-xg73-94fp-g449. - \[Tracking] Fix an issue causing file and model uploads to hang on Databricks ([#8348](https://togithub.com/mlflow/mlflow/issues/8348), [@harupy](https://togithub.com/harupy)) - \[Tracking / Model Registry] Fix an issue causing file and model downloads to hang on Databricks ([#8350](https://togithub.com/mlflow/mlflow/issues/8350), [@dbczumar](https://togithub.com/dbczumar)) - \[Scoring] Fix regression in schema enforcement for model serving when using the `inputs` format for inference ([#8326](https://togithub.com/mlflow/mlflow/issues/8326), [@BenWilson2](https://togithub.com/BenWilson2)) - \[Model Registry] Fix regression in model naming parsing where special characters were not accepted in model names ([#8322](https://togithub.com/mlflow/mlflow/issues/8322), [@arpitjasa-db](https://togithub.com/arpitjasa-db)) - \[Recipes] Fix card rendering with the pandas profiler to handle columns containing all null values ([#8263](https://togithub.com/mlflow/mlflow/issues/8263), [@sunishsheth2009](https://togithub.com/sunishsheth2009)) Documentation updates: - \[Docs] Add an H2O pyfunc usage example to the models documentation ([#8292](https://togithub.com/mlflow/mlflow/issues/8292), [@ericvincent18](https://togithub.com/ericvincent18)) - \[Examples] Add a TensorFlow Core 2.x API usage example ([#8235](https://togithub.com/mlflow/mlflow/issues/8235), [@dheerajnbhat](https://togithub.com/dheerajnbhat)) Small bug fixes and documentation updates: [#8324](https://togithub.com/mlflow/mlflow/issues/8324), [#8325](https://togithub.com/mlflow/mlflow/issues/8325), [@smurching](https://togithub.com/smurching); [#8313](https://togithub.com/mlflow/mlflow/issues/8313), [@dipanjank](https://togithub.com/dipanjank); [#8323](https://togithub.com/mlflow/mlflow/issues/8323), [@liangz1](https://togithub.com/liangz1); [#8331](https://togithub.com/mlflow/mlflow/issues/8331), [#8328](https://togithub.com/mlflow/mlflow/issues/8328), [#8319](https://togithub.com/mlflow/mlflow/issues/8319), [#8316](https://togithub.com/mlflow/mlflow/issues/8316), [#8308](https://togithub.com/mlflow/mlflow/issues/8308), [#8293](https://togithub.com/mlflow/mlflow/issues/8293), [#8289](https://togithub.com/mlflow/mlflow/issues/8289), [#8283](https://togithub.com/mlflow/mlflow/issues/8283), [#8284](https://togithub.com/mlflow/mlflow/issues/8284), [#8285](https://togithub.com/mlflow/mlflow/issues/8285), [#8282](https://togithub.com/mlflow/mlflow/issues/8282), [#8241](https://togithub.com/mlflow/mlflow/issues/8241), [#8270](https://togithub.com/mlflow/mlflow/issues/8270), [#8272](https://togithub.com/mlflow/mlflow/issues/8272), [#8271](https://togithub.com/mlflow/mlflow/issues/8271), [#8268](https://togithub.com/mlflow/mlflow/issues/8268), [@harupy](https://togithub.com/harupy); [#8312](https://togithub.com/mlflow/mlflow/issues/8312), [#8294](https://togithub.com/mlflow/mlflow/issues/8294), [#8295](https://togithub.com/mlflow/mlflow/issues/8295), [#8279](https://togithub.com/mlflow/mlflow/issues/8279), [#8267](https://togithub.com/mlflow/mlflow/issues/8267), [@BenWilson2](https://togithub.com/BenWilson2); [#8290](https://togithub.com/mlflow/mlflow/issues/8290), [@jinzhang21](https://togithub.com/jinzhang21); [#8257](https://togithub.com/mlflow/mlflow/issues/8257), [@WeichenXu123](https://togithub.com/WeichenXu123); [#8307](https://togithub.com/mlflow/mlflow/issues/8307), [@arpitjasa-db](https://togithub.com/arpitjasa-db) ### [`v2.3.0`](https://togithub.com/mlflow/mlflow/blob/HEAD/CHANGELOG.md#230-2023-04-18) [Compare Source](https://togithub.com/mlflow/mlflow/compare/v2.2.2...v2.3.0) MLflow 2.3.0 includes several major features and improvements Features: - \[Models] Introduce a new `transformers` named flavor ([#8236](https://togithub.com/mlflow/mlflow/issues/8236), [#8181](https://togithub.com/mlflow/mlflow/issues/8181), [#8086](https://togithub.com/mlflow/mlflow/issues/8086), [@BenWilson2](https://togithub.com/BenWilson2)) - \[Models] Introduce a new `openai` named flavor ([#8191](https://togithub.com/mlflow/mlflow/issues/8191), [#8155](https://togithub.com/mlflow/mlflow/issues/8155), [@harupy](https://togithub.com/harupy)) - \[Models] Introduce a new `langchain` named flavor ([#8251](https://togithub.com/mlflow/mlflow/issues/8251), [#8197](https://togithub.com/mlflow/mlflow/issues/8197), [@liangz1](https://togithub.com/liangz1), [@sunishsheth2009](https://togithub.com/sunishsheth2009)) - \[Models] Add support for `Pytorch` and `Lightning` 2.0 ([#8072](https://togithub.com/mlflow/mlflow/issues/8072), [@shrinath-suresh](https://togithub.com/shrinath-suresh)) - \[Tracking] Add support for logging LLM input, output, and prompt artifacts ([#8234](https://togithub.com/mlflow/mlflow/issues/8234), [#8204](https://togithub.com/mlflow/mlflow/issues/8204), [@sunishsheth2009](https://togithub.com/sunishsheth2009)) - \[Tracking] Add support for HTTP Basic Auth in the MLflow tracking server ([#8130](https://togithub.com/mlflow/mlflow/issues/8130), [@gabrielfu](https://togithub.com/gabrielfu)) - \[Tracking] Add `search_model_versions` to the fluent API ([#8223](https://togithub.com/mlflow/mlflow/issues/8223), [@mariusschlegel](https://togithub.com/mariusschlegel)) - \[Artifacts] Add support for parallelized artifact downloads ([#8116](https://togithub.com/mlflow/mlflow/issues/8116), [@apurva-koti](https://togithub.com/apurva-koti)) - \[Artifacts] Add support for parallelized artifact uploads for AWS ([#8003](https://togithub.com/mlflow/mlflow/issues/8003), [@harupy](https://togithub.com/harupy)) - \[Artifacts] Add content type headers to artifact upload requests for the `HttpArtifactRepository` ([#8048](https://togithub.com/mlflow/mlflow/issues/8048), [@WillEngler](https://togithub.com/WillEngler)) - \[Model Registry] Add alias support for logged models within Model Registry ([#8164](https://togithub.com/mlflow/mlflow/issues/8164), [#8094](https://togithub.com/mlflow/mlflow/issues/8094), [#8055](https://togithub.com/mlflow/mlflow/issues/8055) [@arpitjasa-db](https://togithub.com/arpitjasa-db)) - \[UI] Add support for custom domain git providers ([#7933](https://togithub.com/mlflow/mlflow/issues/7933), [@gusghrlrl101](https://togithub.com/gusghrlrl101)) - \[Scoring] Add plugin support for customization of MLflow serving endpoints ([#7757](https://togithub.com/mlflow/mlflow/issues/7757), [@jmahlik](https://togithub.com/jmahlik)) - \[Scoring] Add support to MLflow serving that allows configuration of multiple inference workers ([#8035](https://togithub.com/mlflow/mlflow/issues/8035), [@M4nouel](https://togithub.com/M4nouel)) - \[Sagemaker] Add support for asynchronous inference configuration on Sagemaker ([#8009](https://togithub.com/mlflow/mlflow/issues/8009), [@thomasbell1985](https://togithub.com/thomasbell1985)) - \[Build] Remove `shap` as a core dependency of MLflow ([#8199](https://togithub.com/mlflow/mlflow/issues/8199), [@jmahlik](https://togithub.com/jmahlik)) Bug fixes: - \[Models] Fix a bug with `tensorflow` autologging for models with multiple inputs ([#8097](https://togithub.com/mlflow/mlflow/issues/8097), [@jaume-ferrarons](https://togithub.com/jaume-ferrarons)) - \[Recipes] Fix a bug with `Pandas` 2.0 updates for profiler rendering of datetime types ([#7925](https://togithub.com/mlflow/mlflow/issues/7925), [@sunishsheth2009](https://togithub.com/sunishsheth2009)) - \[Tracking] Prevent exceptions from being raised if a parameter is logged with an existing key whose value is identical to the logged parameter ([#8038](https://togithub.com/mlflow/mlflow/issues/8038), [@AdamStelmaszczyk](https://togithub.com/AdamStelmaszczyk)) - \[Tracking] Fix an issue with deleting experiments in the FileStore backend ([#8178](https://togithub.com/mlflow/Configuration
📅 Schedule: Branch creation - "" (UTC), Automerge - At any time (no schedule defined).
🚦 Automerge: Disabled by config. Please merge this manually once you are satisfied.
♻ Rebasing: Whenever PR becomes conflicted, or you tick the rebase/retry checkbox.
🔕 Ignore: Close this PR and you won't be reminded about these updates again.
This PR has been generated by Mend Renovate. View repository job log here.