Open source platform for the machine learning lifecycle
MLflow 2.3.2 is a patch release containing the following features, bug fixes and changes:
Features:
transformers
models pyfunc
inference and serving (#8375, @ankit-db)transformers
model (#8405, @BenWilson2)torch_dtype
values in transformers
pipelines (#8421, @BenWilson2)Feature Extraction
pipelines in the transformers
flavor (#8423, @BenWilson2)Bug Fixes:
Text2TextGeneration
pipelines in the transformers
flavor (#8391, @BenWilson2)Documentation updates:
signature
logging to all examples and documentation (#8410, #8401, #8400, #8387 @jerrylian-db)sentence-transformers
examples to the transformers
examples suite (#8425, @BenWilson2)trubrics
(#8371, @jeffkayne)gluon
pyfunc example to Model flavor documentation (#8403, @ericvincent18)statsmodels
pyfunc example to Models
flavor documentation (#8394, @ericvincent18)Small bug fixes and documentation updates:
#8415, #8412, #8411, #8355, #8354, #8353, #8348, @harupy; #8374, #8367, #8350, @dbczumar; #8358 @mrkaye97; #8392, #8362, @smurching; #8427, #8408, #8399, #8381, @BenWilson2; #8395, #8390, @jerrylian-db; #8402, #8398, @WeichenXu123; #8377, #8363, @arpitjasa-db; #8385, @prithvikannan; #8418, @Jeukoh;
MLflow 2.3.1 is a patch release containing bug fixes and a security patch for https://github.com/mlflow/mlflow/security/advisories/GHSA-83fm-w79m-64r5. If you are using mlflow server
or mlflow ui
, we recommend upgrading to MLflow 2.3.1 as soon as possible.
Security patches:
Bug fixes:
inputs
format for inference (#8326, @BenWilson2)MLflow 2.3.0 includes several major features and improvements
Features:
transformers
named flavor (#8236, #8181, #8086, @BenWilson2)openai
named flavor (#8191, #8155, @harupy)langchain
named flavor (#8251, #8197, @liangz1, @sunishsheth2009)Pytorch
and Lightning
2.0 (#8072, @shrinath-suresh)search_model_versions
to high-level fluent API (#8223, @mariusschlegel)HttpArtifactRepository
(#8048, @WillEngler)shap
as a core dependency of MLflow (#8199, @jmahlik)Bug fixes:
tensorflow
autologging for models with multiple inputs (#8097, @jaume-ferrarons)Pandas
2.0 updates for profiler rendering of datetime types (#7925, @sunishsheth2009)UNFINISHED
(#8154, @WeichenXu123)lightning
hyperparameter tuning examples (#8039, @BenWilson2)Documentation updates:
Small bug fixes and documentation updates:
#8262, #8252, #8250, #8228, #8221, #8203, #8134, #8040, #7994, #7934, @BenWilson2; #8258, #8255, #8253, #8248, #8247, #8245, #8243, #8246, #8244, #8242, #8240, #8229, #8198, #8192, #8112, #8165, #8158, #8152, #8148, #8144, #8143, #8120, #8107, #8105, #8102, #8088, #8089, #8096, #8075, #8073, #8076, #8063, #8064, #8033, #8024, #8023, #8021, #8015, #8005, #7982, #8002, #7987, #7981, #7968, #7931, #7930, #7929, #7917, #7918, #7916, #7914, #7913, @harupy; #7955, @arjundc-db; #8219, #8110, #8093, #8087, #8091, #8092, #8029, #8028, #8031, @jerrylian-db; #8187, @apurva-koti; #8210, #8001, #8000, @arpitjasa-db; #8161, #8127, #8095, #8090, #8068, #8043, #7940, #7924, #7923, @dbczumar; #8147, @morelen17; #8106, @WeichenXu123; #8117, @eltociear; #8100, @laerciop; #8080, @elado; #8070, @grofte; #8066, @yukimori; #8027, #7998, @liangz1; #7999, @martlaf; #7964, @viditjain99; #7928, @alekseyolg; #7909, #7901, #7844, @smurching; #7971, @n30111; #8012, @mingyu89; #8137, @lobrien; #7992, @robmarkcole; #8263, @sunishsheth2009
MLflow 2.2.2 is a patch release containing the following bug fixes:
source
to be a local path within a run's artifact directory if a run_id
is specified (#7993, @harupy)name
to be a file path in FileStore.get_registered_model
(#7965, @harupy)MLflow 2.2.1 is a patch release containing the following bug fixes and security patches:
MlflowClient.search_model_versions()
(#7935, @dbczumar)MLflow 2.2.0 includes several major features and improvements
Features:
model_format
when autologging XGBoost models (#7781, @guyrosin)MLFLOW_ARTIFACT_UPLOAD_DOWNLOAD_TIMEOUT
environment variable to configure artifact operation timeouts (#7783, @wamartin-aml)Content-Type
response headers for artifacts downloaded from mlflow server
(#7827, @bali0019)searchModelVersions()
API to the Java client (#7880, @gabrielfu)max_results
, order_by
and page_token
arguments to MlflowClient.search_model_versions()
(#7623, @serena-ruan)MLFLOW_DEFAULT_PREDICTION_DEVICE
environment variable to set the device for pyfunc model inference (#7922, @ankit-db)Bug fixes:
inspect()
is called (#7852, @sunishsheth2009)positive_class
configuration in the transform step (#7626, @sunishsheth2009)mlflow.evaluate()
(#7613, @sunishsheth2009)run_id
and artifact_path
keys to logged MLmodel files (#7651, @sunishsheth2009)mlflow server
(#7908, @harupy)PYTHONOPTIMIZE=2
(#7791, @dbczumar)Documentation updates:
mlflow.lightgbm
APIs (#7565, @canerturkseven)sktime
(#7624, @benjaminbluhm)precision_recall_auc
metric calculation in mlflow.evaluate()
(#7701, @BenWilson2)Small bug fixes and documentation updates:
#7866, #7751, #7724, #7699, #7697, #7666, @alekseyolg; #7896, #7861, #7858, #7862, #7872, #7859, #7863, #7767, #7766, #7765, #7741, @smurching; #7895, #7877, @viditjain99; #7898, @midhun1998; #7891, #7892, #7886, #7882, #7883, #7875, #7874, #7871, #7868, #7854, #7847, #7845, #7838, #7830, #7837, #7836, #7834, #7831, #7828, #7825, #7826, #7824, #7823, #7778, #7780, #7776, #7775, #7773, #7772, #7769, #7756, #7768, #7764, #7685, #7726, #7722, #7720, #7423, #7712, #7710, #7713, #7688, #7663, #7674, #7673, #7672, #7662, #7653, #7646, #7615, #7614, #7586, #7601, #7598, #7602, #7599, #7577, #7585, #7583, #7584, @harupy; #7865, #7803, #7753, #7719, @dipanjank; #7796, @serena-ruan; #7849, @turbotimon; #7822, #7600, @WeichenXu123; #7811, @guyrosin; #7812, #7788, #7787, #7748, #7730, #7616, #7593, @dbczumar; #7793, @Joel-hanson; #7792, #7694, #7643, @BenWilson2; #7771, #7657, #7644, @nsenno-dbr; #7738, @wkrt7; #7740, @Ark-kun; #7739, #7733, @bali0019; #7723, @andrehp; #7691, #7582, @agoyot; #7721, @Eseeldur; #7709, @srowen; #7693, @ry3s; #7649, @funkypenguin; #7665, @benjaminbluhm; #7668, @eltociear; #7550, @danielhstahl; #7920, @arjundc-db
MLflow 2.1.1 is a patch release containing the following bug fixes:
mlflow.pyfunc.spark_udf()
type casting error on model with ColSpec
input schema
and make PyFuncModel.predict
support dataframe with elements of numpy.ndarray
type (#7592 @WeichenXu123)mlflow.pyfunc.scoring_server.client.ScoringServerClient
support input dataframe with elements
of numpy.ndarray
type (#7594 @WeichenXu123)MLflow 2.1.0 includes several major features and improvements
Features:
/version
endpoint to mlflow server
for querying the server's MLflow version (#7273, @joncarter1)mlflow.search_registered_models()
fluent API (#7428, @TSienki)getRegisteredModel()
method to the Java client (#6602) (#7511, @drod331)mlflow_set_model_version_tag()
method to the R client (#7401, @leeweijie)metadata
field to the MLmodel specification and log_model()
methods (#7237, @jdonzallaz)Model.load()
to support loading MLmodel specifications from remote locations (#7517, @dbczumar)requirements.txt
and conda.yaml
files (#7364, @BenWilson2)mlflow.pyfunc.spark_udf()
to support StructType results (#7527, @WeichenXu123)mlflow.pyfunc.spark_udf()
(#7531, #7291, @WeichenXu123)Bug fixes:
early_stop
functions during model tuning (#7538, @sunishsheth2009)mlflow.autolog()
consistent with mlflow.evaluate()
(#7418, @wenfeiy-db)mlflow.pyfunc.spark_udf()
(#7427, @WeichenXu123)Documentation updates:
dataframe_split
format (#7540, @zhouyangyu)dataframe_records
format (#7361, @dbczumar)Small bug fixes and documentation updates:
#7571, #7543, #7529, #7435, #7399, @WeichenXu123; #7568, @xiaoye-hua; #7549, #7557, #7509, #7498, #7499, #7485, #7486, #7484, #7391, #7388, #7390, #7381, #7366, #7348, #7346, #7334, #7340, #7323, @BenWilson2; #7561, #7562, #7560, #7553, #7546, #7539, #7544, #7542, #7541, #7533, #7507, #7470, #7469, #7467, #7466, #7464, #7453, #7449, #7450, #7440, #7430, #7436, #7429, #7426, #7410, #7406, #7409, #7407, #7405, #7396, #7393, #7395, #7384, #7376, #7379, #7375, #7354, #7353, #7351, #7352, #7350, #7345, #6493, #7343, #7344, @harupy; #7494, @dependabot[bot]; #7526, @tobycheese; #7489, @liangz1; #7534, @Jingnan-Jia; #7496, @danielhstahl; #7504, #7503, #7459, #7454, #7447, @tsugumi-sys; #7461, @wkrt7; #7451, #7414, #7372, #7289, @sunishsheth2009; #7441, @ikrizanic; #7432, @Pochingto; #7386, @jhallard; #7370, #7373, #7371, #7336, #7341, #7342, @dbczumar; #7335, @prithvikannan
The 2.0.1 version of MLflow is a major milestone release that focuses on simplifying the management of end-to-end MLOps workflows, providing new feature-rich functionality, and expanding upon the production-ready MLOps capabilities offered by MLflow. This release contains several important breaking changes from the 1.x API, additional major features and improvements.
Features:
Breaking Changes:
The following list of breaking changes are arranged by their order of significance within each category.
Bug fixes:
Small bug fixes and documentation updates:
#7309, #7314, #7288, #7276, #7244, #7207, #7175, #7107, @sunishsheth2009; #7261, #7313, #7311, #7249, #7278, #7260, #7284, #7283, #7263, #7266, #7264, #7267, #7265, #7250, #7259, #7247, #7242, #7143, #7214, #7226, #7230, #7227, #7229, #7225, #7224, #7223, #7210, #7192, #7197, #7196, #7204, #7198, #7191, #7189, #7184, #7182, #7170, #7183, #7131, #7165, #7151, #7164, #7168, #7150, #7128, #7028, #7118, #7117, #7102, #7072, #7103, #7101, #7100, #7099, #7098, #7041, #7040, #6978, #6768, #6719, #6669, #6658, #6656, #6655, #6538, #6507, #6504 @harupy; #7310, #7308, #7300, #7290, #7239, #7220, #7127, #7091, #6713 @BenWilson2; #7299, #7271, #7209, #7180, #7179, #7158, #7147, #7114, @prithvikannan; #7275, #7245, #7134, #7059, @jinzhang21; #7306, #7298, #7287, #7272, #7258, #7236, @ayushthe1; #7279, @tk1012; #7219, @rddefauw; #7218, #7208, #7188, #7190, #7176, #7137, #7136, #7130, #7124, #7079, #7052, #6541 @dbczumar; #6640, @WeichenXu123; #7200, @hubertzub-db; #7121, @Gonmeso; #6988, @alonisser; #7141, @pdifranc; #7086, @jerrylian-db; #7286, @shogohida