Python SDK Changelog
Updates to the Arize Python SDK
Version 7.11
7.11.1 (Mar 5, 2024)
Fix a bug that caused
ImportError
when importingClient
fromarize.api
7.11.0 (Feb 23, 2024)
Optional strict typing in pandas logger Schema
Optional strict typing in record-at-a-time logger
Add optional extra dependencies if the Arize package is installed as
pip install arize[NLP_Metrics]
:nltk>=3.0.0, <4
sacrebleu>=2.3.1, <3
rouge-score>=0.1.2, <1
evaluate>=0.3, <1
Version 7.10
7.10.2 (Feb 14, 2024)
Check that space and API keys are of string type
Address backward compatibility issue for batch logging via Pandas for on-prem customers
7.10.1 (Feb 6, 2024)
Our Tracing extra requirements now include
deprecated
, a dependency coming fromopentelemetry-semantic-conventions
, which absence produced anImportError
7.10.0 (Feb 1, 2024)
New batch ingestion via Pandas DataFrames for
MULTICLASS
model typeNew
TRACING
environment. You can now log spans & traces for your LLM applications into Arize using batch ingestion via Pandas DataFramesRemoved size limitation on our Schema. You can now log wider models (more columns in your DataFrame)
Prediction ID and Ranking Group ID have an increased character limit from 128 to 512
Our MimicExplainer extra requirements are now more relaxed.
We only require
interpret-community[mimic]>=0.22.0,<1
Version 7.9
7.9.0 (Dec 28, 2023)
New
MULTICLASS
model type available for record-at-a-time ingestion
Version 7.8
7.8.1 (Dec 18, 2023)
Fix a bug that caused missing columns validation feedback to have repeated columns in the message
Fix a bug that caused a
KeyError
whenllm_params
is not found in the dataframe. Improved feedback to the user was included.
7.8.0 (Dec 13, 2023)
Enable latent actuals for
GENERATIVE_LLM
models
Enable feedback when files are too large for better user experience and troubleshooting
Updated
pandas
requirement. We now accept pandas 2.x
Version 7.7
7.7.2 (Nov 9, 2023)
Default prediction sent as string for
GENERATIVE_LLM
single-record-logger (before it was incorrectly set as an integer, resulting in it being categorized as prediction score instead of prediction label)
7.7.1 (Nov 8, 2023)
Check the value of prompt/response raw_data only if not None
7.7.0 (Nov 2, 2023)
Add
CORPUS
supportAccept strings for prompt and response
Make prompt and response optional
Add support for a list of strings features in single-record-logger
Avoid creating a view of a Pandas dataframe
Version 7.6
7.6.1 (Oct 24, 2023)
Add validation on embedding raw data for batch and record-at-a-time loggers
Raise validation string limits for string fields
Add truncation warnings for long string fields
7.6.0 (Oct 12, 2023)
New ability to send features with type
list[str]
New fields available to send token usage to Arize, both using our pandas batch logger and the single record logger
Version 7.5
7.5.1 (Oct 5, 2023)
Increase time interval validation from 2 years to 5 years
Require
python>=3.6
(as opposed topython>=3.8
) for our core SDK. Our extras still requirepython>=3.8
. See Python SDK for more details.Require
pyarrow>=0.15.0
(as opposed topyarrow>=5.0.0
)
7.5.0 (Sep 2, 2023)
Add prompt templates and LLM config fields to the single log and pandas batch ingestion. These fields are used in the Arize Prompt Template Playground
Add a validation check that fails if there are more than 30 embedding features sent
Version 7.4
7.4.0 (Aug 15, 2023)
Add filtering via the keyword
where
to the Exporter client
Version 7.3
7.3.0 (Aug 1, 2023)
AutoEmbeddings supports any model in the HuggingFace Hub, public or private.
Add AutoEmbeddings
UseCase
for Object DetectionAdd
EmbeddingGenerator.list_default_models()
method
Computer Vision AutoEmbeddings switched from using
FeatureExtractor
(deprecated from HuggingFace) toImageProcessor
class
Version 7.2
7.2.0 (Jul 22, 2023)
Authenticating Arize Client using environment variables
A bug causing permission errors for pandas logging using Windows machines
A bug forcing tags to be strings
Version 7.1
7.1.0 (Jun 26, 2023)
Add Generative LLM model-type support for single-record logging
Version 7.0
7.0.6 (Jun 24, 2023)
Removed dependency on
interpret
for the MimicExplainer
7.0.5 (Jun 23, 2023)
Add a progress bar to the Exporter client
Sort exported dataframe by time
Update reserved headers
Add validation check to Exporter client that will fail if
start_time > end_time
Add bug causing to return an error when a query returns no data. Instead, return an empty response
A bug causing the Exporter client to return empty columns in the dataframe if there was no data in them
A bug causing incorrect parsing of
GENERATIVE_LLM
model fields:prompt
&response
Add missing dependency for Exporter:
tqdm>=4.60.0,<5
7.0.4 (Jun 13, 2023)
Relax protobuf requirements from
protobuf~=3.12
toprotobuf>=3.12, <5
7.0.3 (Jun 2, 2023)
Python Export Client, you can now export data from Arize using the Python SDK
A bug preventing
REGRESSION
models from using the MimicExplainer
Remove null value validation for
prediction_label
andactual_label
from single-record loggingAdd model mapping rules validation for
OBJECT_DETECTION
models
7.0.2 (May 12, 2023)
Improve error messages around prediction ID, prediction labels, and tags
A bug causing predictions to be sent as scores instead of labels for
NUMERIC
model types
Add a validation check that will fail if the character limit on tags (1000 max) is exceeded
Add a validation check that will fail if actuals are sent without prediction ID information (for single-record logging). This would result in a delayed record being sent without a prediction ID, which is necessary for the latent join
Add a validation check that will fail if the
Schema
, without prediction columns, does not contain a prediction ID column (for pandas logging). This would result in a delayed record being sent without a prediction ID, which is necessary for the latent joinAdd a validation check that will fail if the
Schema
points to an empty string as a column nameAdd check for invalid index in AutoEmbeddings: DataFrames must have a sorted, continuous index starting at 0
Remove label requirements & accept null values on
SCORE_CATEGORICAL
,NUMERIC
, andRANKING
modelsAllow feature and tag columns to contain null values for pandas logging
Allow to send delayed actuals for
RANKING
models, it is no longer enforced the presence ofrank
andprediction_group_id
columns in theSchema
. However, if the columns are sent, they must not have nulls, since we cannot construct predictions with either value null
Change optional dependency for
MimicExplainer
, raise the version ceiling oflightgbm
from 3.3.4 to 4
7.0.1 (Apr 25, 2023)
A bug causing
GENERATIVE_LLM
models to be sent asSCORE_CATEGORICAL
models
7.0.0 (Apr 13, 2023)
Add Object Detection model-type support
Add Generative LLM model-type support for pandas logging
Add evaluation metrics generation for Generative LLM models
Make prediction IDs optional
Add summarization
UseCase
to AutoEmbeddingsAdd optional, additional custom headers to
Client
instantiation
Add a warning message when only actuals are sent
Add a descriptive error message when embedding features are sent without a vector
Add warning when prediction label or prediction ID will be defaulted
A bug causing skipped validation checks when the new REGRESSION and CATEGORICAL model types are selected
Add a validation check that will fail if the character limit on prediction ID (128 max) is exceeded
Add a validation check that will fail if there are duplicated columns in the dataframe
Changed time range requirements to -2/+1 (two years in the past, and 1 future year)
Require
Python >= 3.8
Add optional extra dependencies if the Arize package is installed as
pip install arize[LLM_Evaluation]
:nltk>=3.0.0, <4
sacrebleu>=2.3.1, <3
rouge-score>=0.1.2, <1
evaluate>=0.3, <1
Remove
numeric_sequence
support
Last updated