Mlflow

Features:

- Added `mlflow.azureml.deploy` API for deploying MLflow models to AzureML (2375 csteegz, 2711, akshaya-a)
- Added support for case-sensitive LIKE and case-insensitive ILIKE queries (e.g. `'params.framework LIKE '%sklearn%'`) with the SearchRuns API & UI when running against a SQLite backend (2217, t-henri; 2708, mparkhe)
- Improved line smoothing in MLflow metrics UI using exponential moving averages (2620, Valentyn1997)
- Added `mlflow.spacy` module with support for logging and loading spaCy models (2242, arocketman)
- Parameter values that differ across runs are highlighted in run comparison UI (2565, gabrielbretschner)
- Added ability to compare source runs associated with model versions from the registered model UI (2537, juntai-zheng)
- Added support for alphanumerical experiment IDs in the UI. (2568, jonas)
- Added support for passing arguments to `docker run` when running docker-based MLflow projects (2608, ksanjeevan)
- Added Windows support for `mlflow sagemaker build-and-push-container` CLI & API (2500, AndreyBulezyuk)
- Improved performance of reading experiment data from local filesystem when LibYAML is installed (2707, Higgcz)
- Added a healthcheck endpoint to the REST API server at `/health` that always returns a 200 response status code, to be used to verify health of the server (2725, crflynn)
- MLflow metrics UI plots now scale to rendering thousands of points using scattergl (2447, mjlbach)

Bug fixes:

- Fixed CLI summary message in `mlflow azureml build_image` CLI (2712, dbczumar)
- Updated `examples/flower_classifier/score_images_rest.py` with multiple bug fixes (2647, tfurmston)
- Fixed pip not found error while packaging models via `mlflow models build-docker` (2699, HiromuHota)
- Fixed bug in `mlflow.tensorflow.autolog` causing erroneous deletion of TensorBoard logging directory (2670, dbczumar)
- Fixed a bug that truncated the description of the `mlflow gc` subcommand in `mlflow --help` (2679, dbczumar)
- Fixed bug where `mlflow models build-docker` was failing due to incorrect Miniconda download URL (2685, michaeltinsley)
- Fixed a bug in S3 artifact logging functionality where `MLFLOW_S3_ENDPOINT_URL` was ignored (2629, poppash)
- Fixed a bug where Sqlite in-memory was not working as a tracking backend store by modifying DB upgrade logic (2667, dbczumar)
- Fixed a bug to allow numerical parameters with values >= 1000 in R `mlflow::mlflow_run()` API (2665, lorenzwalthert)
- Fixed a bug where AWS creds was not found in the Windows platform due path differences (2634, AndreyBulezyuk)
- Fixed a bug to add pip when necessary in `_mlflow_conda_env` (2646, tfurmston)
- Fixed error code to be more meaningful if input to model version is incorrect (2625, andychow-db)
- Fixed multiple bugs in model registry (2638, aarondav)
- Fixed support for conda env dicts with `mlflow.pyfunc.log_model` (2618, dbczumar)
- Fixed a bug where hiding the start time column in the UI would also hide run selection checkboxes (2559, harupy)

Documentation updates:

- Added links to source code to mlflow.org (2627, harupy)
- Documented fix for pandas-records payload (2660, SaiKiranBurle)
- Fixed documentation bug in TensorFlow `load_model` utility (2666, pogil)
- Added the missing Model Registry description and link on the first page (2536, dmatrix)
- Added documentation for expected datatype for step argument in `log_metric` to match REST API (2654, mparkhe)
- Added usage of the model registry to the `log_model` function in `sklearn_elasticnet_wine/train.py` example (2609, netanel246)

Small bug fixes and doc updates (2594, Trollgeir; 2703,2709, juntai-zheng; 2538, 2632, keigohtr; 2656, 2553, lorenzwalthert; 2622, pingsutw; 2615, 2600, 2533, mlflow-automation; 1391, sueann; 2613, 2598, 2534, 2723, smurching; 2652, 2710, mparkhe; 2706, 2653, 2639, tomasatdatabricks; 2611, 9dogs; 2700, 2705, aarondav; 2675, 2540, mengxr; 2686, RensDimmendaal; 2694, 2695, 2532, dbczumar; 2733, 2716, harupy; 2726, crflynn; 2582, 2687, dmatrix)

- Pin alembic version to 1.4.1 or below to prevent pep517-related installation errors
(2612, smurching)

- Remove usage of Nonnull annotations and findbugs dependency in Java package (2583, mparkhe)
- Add version upper bound (<=1.3.13) to sqlalchemy dependency in Python package (2587, smurching)

Other bugfixes and doc updates (2595, mparkhe; 2567, jdlesage)

MLflow support for Python 2 is now deprecated and will be dropped in a future release. At that
point, existing Python 2 workflows that use MLflow will continue to work without modification, but
Python 2 users will no longer get access to the latest MLflow features and bugfixes. We recommend
that you upgrade to Python 3 - see https://docs.python.org/3/howto/pyporting.html for a migration
guide.

Breaking changes to Model Registry REST APIs:

Model Registry REST APIs have been updated to be more consistent with the other MLflow APIs. With
this release Model Registry APIs are intended to be stable until the next major version.

- Python and Java client APIs for Model Registry have been updated to use the new REST APIs. When using an MLflow client with a server using updated REST endpoints, you won't need to change any code but will need to upgrade to a new client version. The client APIs contain deprecated arguments, which for this release are backward compatible, but will be dropped in future releases. (2457, tomasatdatabricks; 2502, mparkhe).
- The Model Registry UI has been updated to use the new REST APIs (2476 aarondav; 2507, mparkhe)

Other Features:

- Ability to click through to individual runs from metrics plot (2295, harupy)
- Added `mlflow gc` CLI for permanent deletion of runs (2265, t-henri)
- Metric plot state is now captured in page URLs for easier link sharing (2393, 2408, 2498 smurching; 2459, harupy)
- Added experiment management to MLflow UI (create/rename/delete experiments) (2348, ggliem)
- Ability to search for experiments by name in the UI (2324, ggliem)
- MLflow UI page titles now reflect the content displayed on the page (2420, AveshCSingh)
- Added a new `LogModel` REST API endpoint for capturing model metadata, and call it from the Python and R clients (2369, 2430, 2468 tomasatdatabricks)
- Java Client API to download model artifacts from Model Registry (2308, andychow-db)

Bug fixes and documentation updates:

- Updated Model Registry documentation page with code snippets and examples (2493, dmatrix; 2517, harupy)
- Better error message for Model Registry, when using incompatible backend server (2456, aarondav)
- matplotlib is no longer required to use XGBoost and LightGBM autologging (2423, harupy)
- Fixed bug where matplotlib figures were not closed in XGBoost and LightGBM autologging (2386, harupy)
- Fixed parameter reading logic to support param values with newlines in FileStore (2376, dbczumar)
- Improve readability of run table column selector nodes (2388, dbczumar)
- Validate experiment name supplied to `UpdateExperiment` REST API endpoint (2357, ggliem)
- Fixed broken MLflow DB README link in CLI docs (2377, dbczumar)
- Change copyright year across docs to 2020 (2349, ParseThis)

Small bug fixes and doc updates (2378, 2449, 2402, 2397, 2391, 2387, 2523, 2527 harupy; 2314, juntai-zheng; 2404, andychow-db; 2343, pogil; 2366, 2370, 2364, 2356, AveshCSingh; 2373, 2365, 2363, smurching; 2358, jcuquemelle; 2490, RensDimmendaal; 2506, dbczumar; 2234 Zangr; 2359 lbernickm; 2525, mparkhe)

Features:

- Adds a new runs table column view based on `ag-grid` which adds functionality for nested runs, serverside sorting, column reordering, highlighting, and more. (2251, Zangr)
- Adds contour plot to the run comparsion page to better support parameter tuning (2225, harupy)
- If you use EarlyStopping with Keras autologging, MLflow now automatically captures the best model trained and the associated metrics (2301, 2219, juntai-zheng)
- Adds autologging functionality for LightGBM and XGBoost flavors to log feature importance, metrics per iteration, the trained model, and more. (2275, 2238, harupy)
- Adds an experimental mlflow.spark.autolog() API for automatic logging of Spark datasource information to the current active run. (2220, smurching)
- Optimizes the file store to load less data from disk for each operation (2339, jonas)
- Upgrades from ubuntu:16.04 to ubuntu:18.04 when building a Docker image with `mlflow models build-docker` (2256, andychow-db)

Bug fixes and documentation updates:

- Fixes bug when running server against database URLs with symbols (2289, hershaw)
- Fixes model Docker image build on Windows (2257, jahas)
- Documents the SQL Server plugin (2320, avflor)
- Adds a help file for the R package (2259, lorenzwalthert)
- Adds an example of using the Search API to find the best performing model (2313, AveshCSingh)
- Documents how to write and use MLflow plugins (2270, smurching)

Small bug fixes and doc updates (2293, 2328, 2244, harupy; 2269, 2332, 2306, 2307, 2292, 2267, 2191, 2231, juntai-zheng; 2325, shubham769; 2291, sueann; 2315, 2249, 2288, 2278, 2253, 2181, smurching; 2342, tomasatdatabricks; 2245, dependabot[bot]; 2338, jcuquemelle; 2285, avflor; 2340, pogil; 2237, 2226, 2243, 2272, 2286, dbczumar; 2281, renaudhager; 2246, avaucher; 2258, lorenzwalthert; 2261, smith-kyle; 2352, dbczumar)

New Model Flavors and Flavor Updates:

- New support for a LightGBM flavor (2136, harupy)
- New support for a XGBoost flavor (2124, harupy)
- New support for a Gluon flavor and autologging (1973, cosmincatalin)
- Runs automatically created by `mlflow.tensorflow.autolog()` and `mlflow.keras.autolog()` (2088) are now automatically ended after training and/or exporting your model. See the [`docs`](https://mlflow.org/docs/latest/tracking.html#automatic-logging-from-tensorflow-and-keras-experimental) for more details (2094, juntai-zheng)

More features and improvements:

- When using the `mlflow server` CLI command, you can now expose metrics on `/metrics` for Prometheus via the optional --activate-parameter argument (2097, t-henri)
- The `mlflow ui` CLI command now has a `--host`/`-h` option to specify user-input IPs to bind to (2176, gandroz)
- MLflow now supports pulling Git submodules while using MLflow Projects (2103, badc0re)
- New `mlflow models prepare-env` command to do any preparation necessary to initialize an environment. This allows distinguishing configuration and user errors during predict/serve time (2040, aarondav)
- TensorFlow.Keras and Keras parameters are now logged by `autolog()` (2119, juntai-zheng)
- MLflow `log_params()` will recognize Spark ML params as keys and will now extract only the name attribute (2064, tomasatdatabricks)
- Exposes `mlflow.tracking.is_tracking_uri_set()` (2026, fhoering)
- The artifact image viewer now displays "Loading..." when it is loading an image (1958, harupy)
- The artifact image view now supports animated GIFs (2070, harupy)
- Adds ability to mount volumes and specify environment variables when using mlflow with docker (1994, nlml)
- Adds run context for detecting job information when using MLflow tracking APIs within Databricks Jobs. The following job types are supported: notebook jobs, Python Task jobs (2205, dbczumar)
- Performance improvement when searching for runs (2030, 2059, jcuquemelle; 2195, rom1504)

Bug fixes and documentation updates:

- Fixed handling of empty directories in FS based artifact repositories (1891, tomasatdatabricks)
- Fixed `mlflow.keras.save_model()` usage with DBFS (2216, andychow-db)
- Fixed several build issues for the Docker image (2107, jimthompson5802)
- Fixed `mlflow_list_artifacts()` (R package) (2200, lorenzwalthert)
- Entrypoint commands of Kubernetes jobs are now shell-escaped (2160, zanitete)
- Fixed project run Conda path issue (2147, Zangr)
- Fixed spark model load from model repository (2175, tomasatdatabricks)
- Stripped "dev" suffix from PySpark versions (2137, dbczumar)
- Fixed note editor on the experiment page (2054, harupy)
- Fixed `models serve`, `models predict` CLI commands against models:/ URIs (2067, smurching)
- Don't unconditionally format values as metrics in generic HtmlTableView component (2068, smurching)
- Fixed remote execution from Windows using posixpath (1996, aestene)
- Add XGBoost and LightGBM examples (2186, harupy)
- Add note about active run instantiation side effect in fluent APIs (2197, andychow-db)
- The tutorial page has been refactored to be be a 'Tutorials and Examples' page (2182, juntai-zheng)
- Doc enhancements for XGBoost and LightGBM flavors (2170, harupy)
- Add doc for XGBoost flavor (2167, harupy)
- Updated `active_run()` docs to clarify it cannot be used accessing current run data (2138, juntai-zheng)
- Document models:/ scheme for URI for load_model methods (2128, stbof)
- Added an example using Prophet via pyfunc (2043, dr3s)
- Added and updated some screenshots and explicit steps for the model registry (2086, stbof)

Small bug fixes and doc updates (2142, 2121, 2105, 2069, 2083, 2061, 2022, 2036, 1972, 2034, 1998, 1959, harupy; 2202, t-henri; 2085, stbof; 2098, AdamBarnhard; 2180, 2109, 1977, 2039, 2062, smurching; 2013, aestene; 2146, joelcthomas; 2161, 2120, 2100, 2095, 2088, 2076, 2057, juntai-zheng; 2077, 2058, 2027, sueann; 2149, zanitete; 2204, 2188, andychow-db; 2110, 2053, jdlesage; 2003, 1953, 2004, Djailla; 2074, nlml; 2116, Silas-Asamoah; 1104, jimthompson5802; 2072, cclauss; 2221, 2207, 2157, 2132, 2114, 2063, 2065, 2055, dbczumar; 2033, cthoyt; 2048, philip-khor; 2002, jspoorta; 2000, christang; 2078, dennyglee; 1986, vguerra; 2020, dependabot[bot])