v1.1.44

whylogs release v1.1.44

Hi everyone! We’ve now released whylogs 1.1.44 🚀.
whylogs is the open standard for data and ML logging created by WhyLabs. 👩🏽‍🔬
This version includes:

• Fix UDF deserialization in column profiles to support UDFs on pyspark #1271 [@jamie256]
• Add schema namespaces for UDFs #1268 [@richard-rogers]
• Add optional name parameter to log methods #1263 [@jamie256]
• Avoid empty ints series cast errors on timestamps #1265 [@jamie256]
• Support for UDF namespaces #1261 [@richard-rogers]

v1.1.43

whylogs release v1.1.43

Hi everyone! We’ve now released whylogs 1.1.43 🚀.
whylogs is the open standard for data and ML logging created by WhyLabs. 👩🏽‍🔬
This version includes:

• Allow hiding condition count metric from whylabs profile #1247 [@richard-rogers]
• Alternate configurable MetricConfig defaults #1257 [@richard-rogers]
• Add an example image logging with udf metric in WhyLabs example #1200 [@jamie256]
• Bump version to release version 1.1.42 #1253 [@jamie256]

✨ Features

• Add factory constraints: condition count below #1256 [@FelipeAdachi]
• Multi-column UDF support #1162 [@richard-rogers]

🐛 Bug Fixes

• change alias to None if empty string #1255 [@FelipeAdachi]

v1.1.42

whylogs release v1.1.42

Hi everyone! We’ve now released whylogs 1.1.42 🚀.
whylogs is the open standard for data and ML logging created by WhyLabs. 👩🏽‍🔬
This version includes:

• Upload segmented reference profile API #1248 [@FelipeAdachi]
• constraints generator fixes - customizable hll_stddev and error handling #1242 [@FelipeAdachi]

v1.1.41

whylogs release v1.1.41

Hi everyone! We’ve now released whylogs 1.1.41 🚀.
whylogs is the open standard for data and ML logging created by WhyLabs. 👩🏽‍🔬
This version includes:

• udf_metric_schema() supports all schema arguments #1237 [@richard-rogers]
• Better handling of exceptions from UDFs

v1.1.40

whylogs release v1.1.40

Hi everyone! We’ve now released whylogs 1.1.40 🚀.
whylogs is the open standard for data and ML logging created by WhyLabs. 👩🏽‍🔬
This version includes:

• Remove unused doc dependencies and restrict docs python version #1233 [@jamie256]
• Improved generate UDF schema #1231 [@richard-rogers]
• Bump version to release version 1.1.39 #1229 [@jamie256]

v1.1.39

whylogs release v1.1.39

Hi everyone! We’ve now released whylogs 1.1.39 🚀.
whylogs is the open standard for data and ML logging created by WhyLabs. 👩🏽‍🔬
This version includes:

• Add a simplified DESCRIPTION.md for pypi's project description #1227 [@jamie256]
• Bump version to release version 1.1.38 #1220 [@jamie256]

v1.1.38

whylogs release v1.1.38

Hi everyone! We’ve now released whylogs 1.1.38 🚀.
whylogs is the open standard for data and ML logging created by WhyLabs. 👩🏽‍🔬
This version includes:

• Add declarative submetric schema for UDF metric #1204 [@richard-rogers]
• Avoid irrelevant constraints generation when column only has null values #1219 [@FelipeAdachi]
• custom quantiles from DistributionMetric #1215 [@jamie256]
• Support setting timestamps on ViewResultSet #1214 [@jamie256]
• Auto constraints expansion #1211 [@FelipeAdachi]
• Bump version to release version 1.1.37 #1212 [@jamie256]

v1.1.37

whylogs release v1.1.37

Hi everyone! We’ve now released whylogs 1.1.37 🚀.
whylogs is the open standard for data and ML logging created by WhyLabs. 👩🏽‍🔬
This version includes:

• Stub out scikit-learn in embeddings metric #1206 [@richard-rogers]
• Fix: removing whylabs_client from extras #1203 [@murilommen]
• Adding whylabs_client as a core dependency #1202 [@murilommen]
• Fix whylabs config read error handling #1201 [@jamie256]

v1.1.36

whylogs release v1.1.36

Hi everyone! We’ve now released whylogs 1.1.36 🚀.
whylogs is the open standard for data and ML logging created by WhyLabs. 👩🏽‍🔬
This version includes:

• Example - Embedding/Tokens/Multi-class tutorial for text data #1181 [@FelipeAdachi]
• Return whylabs profile id on upload #1196 [@jamie256]
• Split bump_mainline workflow #1189 [@jamie256]
• Add image entropy metric #1193 [@jamie256]
• Single value line for double_histogram viz #1192 [@jamie256]
• add whylogs[all] install extra and update poetry.lock #1191 [@jamie256]
• Add preprocessed column entry point for homogeneous columns #1138 [@richard-rogers]
• Experimental UdfMetric #1161 [@richard-rogers]
• Simplify DeclarativeSchema API for adding Condition Count Metrics #1158 [@FelipeAdachi]
• Adds check if cardinality metric exists when getting frequent stats #1186 [@FelipeAdachi]
• Add CardinalityMetric to embeddings submetrics #1187 [@richard-rogers]

🐛 Bug Fixes

• Update to open template file in read-only mode. #1195 [@aq-ashwin]

v1.1.35

whylogs release v1.1.35

Hi everyone! We’ve now released whylogs 1.1.35 🚀.
whylogs is the open standard for data and ML logging created by WhyLabs. 👩🏽‍🔬
This version includes:

• Fix bump2version command in release workflow #1183 [@jamie256]
• Fix back compat deserializing nan/inf counts in ColumnCountsMetric #1185 [@jamie256]