profile
viewpoint
If you are wondering where the data of this site comes from, please visit https://api.github.com/users/mikek/events. GitMemory does not store any data, but only uses NGINX to cache data for a period of time. The idea behind GitMemory is simply to give users a better reading experience.
Mykhailo Kolesnyk mikek Kyiv, Ukraine

mikek/behave-http 40

BDD HTTP steps implementation for Behave

mikek/django-kitchen 3

A knife-solo based Chef kitchen for a Django project provisoning

mikek/django-toplevel-files 1

Simple Django app to manage a few admin-uploaded files

mikek/boto 0

Python interface to Amazon Web Services

mikek/couchdb-cookbook 0

Development repository for Opscode Cookbook couchdb

mikek/crux-mike 0

Personal ports for CRUX linux distribution

mikek/django-basic-project-template 0

Custom Django project template

mikek/django-ckeditor 0

Django admin CKEditor integration.

mikek/django-contact-form 0

Contact form for django 1.3

mikek/django-faq 0

a simple faq application for django apps + some fixes

Pull request review commentdatarobot/datarobot-user-models

[RAPTOR-5302] contains missing validation

 logger = logging.getLogger("drum." + __name__)  -class DataTypes(object):+class Conditions:+    """All acceptable values for the 'condition' field."""++    EQUALS = "EQUALS"+    IN = "IN"+    NOT_EQUALS = "NOT_EQUALS"+    NOT_IN = "NOT_IN"+    GREATER_THAN = "GREATER_THAN"+    LESS_THAN = "LESS_THAN"+    NOT_GREATER_THAN = "NOT_GREATER_THAN"+    NOT_LESS_THAN = "NOT_LESS_THAN"+++class Values:+    """All acceptable values for the 'value' field. """++    NUM = "NUM"+    TXT = "TXT"+    CAT = "CAT"+    IMG = "IMG"+    DATE = "DATE"+    FORBIDDEN = "FORBIDDEN"+    SUPPORTED = "SUPPORTED"+    REQUIRED = "REQUIRED"+    NEVER = "NEVER"+    DYNAMIC = "DYNAMIC"+    ALWAYS = "ALWAYS"+    IDENTITY = "IDENTITY"+++class BaseValidator(object):+    FIELD = None+    CONDITIONS = None+    VALUE = None

VALUES

scottp-ml

comment created time in 3 minutes

Pull request review commentdatarobot/datarobot-user-models

[RAPTOR-5302] contains missing validation

 logger = logging.getLogger("drum." + __name__)  -class DataTypes(object):+class Conditions:+    """All acceptable values for the 'condition' field."""++    EQUALS = "EQUALS"+    IN = "IN"+    NOT_EQUALS = "NOT_EQUALS"+    NOT_IN = "NOT_IN"+    GREATER_THAN = "GREATER_THAN"+    LESS_THAN = "LESS_THAN"+    NOT_GREATER_THAN = "NOT_GREATER_THAN"+    NOT_LESS_THAN = "NOT_LESS_THAN"+++class Values:+    """All acceptable values for the 'value' field. """++    NUM = "NUM"+    TXT = "TXT"+    CAT = "CAT"+    IMG = "IMG"+    DATE = "DATE"+    FORBIDDEN = "FORBIDDEN"+    SUPPORTED = "SUPPORTED"+    REQUIRED = "REQUIRED"+    NEVER = "NEVER"+    DYNAMIC = "DYNAMIC"+    ALWAYS = "ALWAYS"+    IDENTITY = "IDENTITY"+++class BaseValidator(object):+    FIELD = None+    CONDITIONS = None+    VALUE = None++    def __init__(self, condition, values):+        self.condition = condition+        self.values = values++    @classmethod+    def get_yaml_validator(cls):+        return Map(+            {+                "field": Enum(cls.FIELD),+                "condition": Enum(cls.CONDITIONS),+                "value": Enum(cls.VALUES),+            }+        )++    def validate(self, dataframe: pd.DataFrame):+        raise NotImplementedError+++class DataTypes(BaseValidator):     """Validation related to data types.  This is common between input and output."""      FIELD = "data_types"-    VALUES = ["NUM", "TXT", "CAT", "IMG", "DATE"]-    CONDITIONS = ["EQUALS", "IN", "NOT_EQUALS", "NOT_IN"]+    VALUES = [Values.NUM, Values.TXT, Values.CAT, Values.IMG, Values.DATE]+    CONDITIONS = [Conditions.EQUALS, Conditions.IN, Conditions.NOT_EQUALS, Conditions.NOT_IN]     _TYPES = Enum(VALUES)     _DTYPE_MAPPING = {}

may as well delete this

scottp-ml

comment created time in 3 minutes

Pull request review commentdatarobot/datarobot-user-models

[RAPTOR-5302] contains missing validation

 import pandas as pd  +class Conditions:

can/should these be actual Enums?

scottp-ml

comment created time in 26 minutes

pull request commentdatarobot/datarobot-user-models

[RAPTOR-5302] contains missing validation

do tests for custom_model_runner.datarobot_drum.drum.typeschema_validation.revalidate_typeschema need to be updated to test the get_yaml_validator?

scottp-ml

comment created time in 27 minutes

push eventdatarobot/datarobot-user-models

Scott Powers

commit sha 4099844a979fe038c9d3b66dd497f8f61cfeb511

[RAPTOR-5194 RAPTOR-5195]fit time validation (#325) * Adds validation of typeschema yaml format. * Validation schema documentation. * test datatype schema * test schema * incorrect import * perform schema validation at during fit or push * add tests * black * pr comments * use logger * minor fixes * share loading of input file

view details

Scott Powers

commit sha 41bac9cb1e9b723895bfef1ff5b7859c93ac6033

Merge branch 'master' into spp/contains_missing_validation

view details

push time in 3 hours

issue commenthttpie/httpie

brew install httpie will install python too instead of using my existing pyhton

@qiulang Homebrew puts python3 on your $PATH but leaves the version-less python intact so you can use to refer to the system Python. This has nothing to do with HTTPie, though — https://docs.brew.sh/Homebrew-and-Python.

qiulang

comment created time in 4 hours

issue commenthttpie/httpie

brew install httpie will install python too instead of using my existing pyhton

Hi I reproduced the case, please check:

lang@localhost ~ % which python3
/usr/bin/python3
lang@localhost ~ % python3 --version
Python 3.8.2
lang@localhost ~ % brew list
==> Formulae
autoconf	m4		node-build	openssl@1.1	pyenv		sqlite
gdbm		mpdecimal	nodenv		pkg-config	readline	xz
lang@localhost ~ % brew install httpie
...
==> Caveats
==> python@3.9
Python has been installed as
 /usr/local/bin/python3

Then

lang@localhost ~ % python3 --version
Python 3.9.2
//I think it is because the PATH setting
lang@localhost ~ % echo $PATH
/usr/local/bin:/usr/bin:/bin:/usr/sbin:/sbin
qiulang

comment created time in 4 hours

issue commenthttpie/httpie

shadow file name 'ssl.py' may cause run time error when tring to run the program from source code

Yes, I turned out to find my environment may cause the issue, I set up an venv and this problem did not appear again. And python -m httpie.__main__ can properly invoke the program.

zwh-china

comment created time in 10 hours

issue closedhttpie/httpie

shadow file name 'ssl.py' may cause run time error when tring to run the program from source code

image

As you can see in the screenshot in Pycharm, it shows that the ssl has no property named 'CERT_REQUIRED'. But when I looked up the official doc of urlib and pyopenssl. I am sure that it should not be that property doesn't exist.

I reinstall the requirements and openssl plus urlib in the venv still not work. Then I think maybe the shadow name of ssl.py cause the problem. After rename the ssl.py everything went normal.

So I just change the name of ssl.py in the httpie folder and change the import name of ssl in the httpie folder to solve this

closed time in 10 hours

zwh-china

issue commenthttpie/httpie

brew install httpie will install python too instead of using my existing pyhton

Hi macOS 11.2.3 has upgraded system python3 to 3.9 but I will try to find another mac to retry it.

qiulang

comment created time in 15 hours

Pull request review commentdatarobot/datarobot-user-models

[RAPTOR-5203] Add support for Julia Inference Models

+# This is the default base image for use with user models and workflows.+# It contains a variety of common useful data-science packages and tools.+FROM datarobot/python3-dropin-env-base++# Install the list of core requirements, e.g. sklearn, numpy, pandas, flask.+# **Don't modify this file!**+COPY dr_requirements.txt dr_requirements.txt

This was a copy and paste of existing sklearn dropin. Should it be changed in all?

timsetsfire

comment created time in 18 hours

Pull request review commentdatarobot/datarobot-user-models

[RAPTOR-5203] Add support for Julia Inference Models

+# This is the default base image for use with user models and workflows.+# It contains a variety of common useful data-science packages and tools.+FROM datarobot/python3-dropin-env-base++# Install the list of core requirements, e.g. sklearn, numpy, pandas, flask.+# **Don't modify this file!**+COPY dr_requirements.txt dr_requirements.txt++# '--upgrade-strategy eager' will upgrade installed dependencies+# according to package requirements or to the latest+RUN pip3 install -U --upgrade-strategy eager --no-cache-dir --prefer-binary -r dr_requirements.txt  && \+    rm -rf dr_requirements.txt++# Install the list of custom Python requirements, e.g. keras, xgboost, etc.+COPY requirements.txt requirements.txt+RUN pip3 install -r requirements.txt --no-cache-dir && \+    rm -rf requirements.txt++# get and install  julia 1.5.4++ENV HOME /opt+RUN apt-get update && apt-get install wget -y && \+    wget https://julialang-s3.julialang.org/bin/linux/x64/1.5/julia-1.5.4-linux-x86_64.tar.gz && \+    tar zxvf julia-1.5.4-linux-x86_64.tar.gz && \+    mkdir $HOME/.julia && \+    mkdir $HOME/julia++ENV JULIA_HOME /julia-1.5.4+ENV JULIA_SYS_IMAGE $HOME/julia/sys.so+ENV JULIA_PROJECT $HOME/julia+ENV PATH $PATH:$JULIA_HOME/bin+COPY sysim.jl $HOME/julia/sysim.jl+COPY . /opt/code+ENV PYTHON /usr/local/bin/python +RUN julia "$HOME/julia/sysim.jl" && \+    chmod -R 777 $HOME/.julia/logs/manifest_usage.toml && \+    rm julia-1.5.4-linux-x86_64.tar.gz++ENV CODE_DIR=/opt/code +ENV ADDRESS=0.0.0.0:8080 WITH_ERROR_SERVER=1++#Uncomment the following line to switch from Flask to uwsgi server+ENV PRODUCTION=1 MAX_WORKERS=1 SHOW_STACKTRACE=1

FIXED

timsetsfire

comment created time in 18 hours

Pull request review commentdatarobot/datarobot-user-models

[RAPTOR-5203] Add support for Julia Inference Models

+numpy>=1.16.0,<1.19.0+pandas==1.1.0

changed to 1.0.5 per comment.

timsetsfire

comment created time in 18 hours

Pull request review commentdatarobot/datarobot-user-models

[RAPTOR-5203] Add support for Julia Inference Models

+pyarrow==0.14.1+datarobot-drum==1.4.16

FIXED

timsetsfire

comment created time in 18 hours

push eventdatarobot/datarobot-user-models

Scott Powers

commit sha 4099844a979fe038c9d3b66dd497f8f61cfeb511

[RAPTOR-5194 RAPTOR-5195]fit time validation (#325) * Adds validation of typeschema yaml format. * Validation schema documentation. * test datatype schema * test schema * incorrect import * perform schema validation at during fit or push * add tests * black * pr comments * use logger * minor fixes * share loading of input file

view details

Timothy Whittaker

commit sha 50dee7679dc8a3813297cc6a7f93678754adb5c0

[RAPTOR-5203] add julia support for custom inference. increase timeout in DrumServerRun from 10 sec to 30 sec update readmes add julia dropin add julia_predictor.py and necessary components add examples to model_templates/inference black modify run-drum-tests-in-container.sh to build julia sys image for tests

view details

Timothy Whittaker

commit sha 54ef6b8ec1091b88c0f06e12b545b8e7a36a8d14

fix custom_model_runner/requirements.txt and setup.py

view details

Timothy Whittaker

commit sha 2de5c8c804f1053bed3c66c1ee649ec768c49b20

resolving some comments

view details

push time in 18 hours

Pull request review commentdatarobot/datarobot-user-models

[RAPTOR-5203] Add support for Julia Inference Models

+module Custom++using PyCall +export load_model, unstructured_predict++function to_bytes(n::Integer; bigendian=true, len=sizeof(n))+    bytes = Array{UInt8}(undef, len)+    for byte in (bigendian ? (1:len) : reverse(1:len))+        bytes[byte] = n & 0xff+        n >>= 8+    end+    return bytes+end++function load_model(code_dir)  +    return nothing+end++function score_unstructured(data; model = nothing, kwargs = nothing)+    println("Model: $model")+    println("Incoming content type params: $kwargs")+    println("Incoming data type $(typeof(data))")+    println("Incoming data: $data")+    println("Incoming query params: $(kwargs.query)")+    +    if haskey(kwargs.query, "ret_mode")+        ret_mode = kwargs.query["ret_mode"]+    else +        ret_mode = nothing+    end++    word_count = split( replace( data, "\n" => " "), " ")+    word_count = length(word_count)++    if ret_mode == "binary"  +        ret_data = to_bytes(word_count)+        ret_kwargs = PyCall.PyDict( Dict("mimetype" => "application/octet-stream"))+        ret = PyCall.pybytes(ret_data), ret_kwargs+    else +        ret_kwargs = PyCall.PyDict( Dict("mimetype" => kwargs.mimetype)) +        ret = string(word_count)+        #, ret_kwargs+    end+    println("outgoing")+    println(ret)+    return ret+end++end++# print("Model: ", model)

removed

timsetsfire

comment created time in 19 hours

Pull request review commentdatarobot/datarobot-user-models

[RAPTOR-5203] Add support for Julia Inference Models

+## Flux Model ++Source the appropriate Julia Environment++`export JULIA_PROJECT=/Users/timothy.whittaker/Desktop/mlops-experiments/python-julia-sys-image`++`julia --project=$JULIA_PROJECT`++Install `Flux` and `BSON` in your Julia environment if not already available++Again, this is a slow startup time.  For scoring and serving it is fine, but it will always take some time to start everything up, therefore the recommendation is to use a system image of your environment to speed up things considerably.  ++## Scoring++drum score --code-dir model_templates/inference/julia/jl_custom --target-type regression --input tests/testdata/boston_housing_inference.csv --verbose --logging-level info++## Serving with Docker++drum server --code-dir model_templates/inference/julia/jl_custom --target-type regression --input tests/testdata/boston_housing_inference.csv --verbose --logging-level info++++++

on it 👍

timsetsfire

comment created time in 19 hours

Pull request review commentdatarobot/datarobot-user-models

[RAPTOR-5203] Add support for Julia Inference Models

 Include any necessary hooks in a file called `custom.py` for Python models or `c Type checking methods can be used to verify types: - in Python`isinstance(data, str)` or `isinstance(data, bytes)` - in R `is.character(data)` or `is.raw(data)`+- in Julia `data isa String` or `data is Base.CodeUnits`

oh :)

timsetsfire

comment created time in 19 hours

Pull request review commentdatarobot/datarobot-user-models

[RAPTOR-5203] Add support for Julia Inference Models

 Include any necessary hooks in a file called `custom.py` for Python models or `c Type checking methods can be used to verify types: - in Python`isinstance(data, str)` or `isinstance(data, bytes)` - in R `is.character(data)` or `is.raw(data)`+- in Julia `data isa String` or `data is Base.CodeUnits`

the command in julia is isa.

timsetsfire

comment created time in 19 hours

Pull request review commentdatarobot/datarobot-user-models

[RAPTOR-5203] Add support for Julia Inference Models

+# This is the default base image for use with user models and workflows.+# It contains a variety of common useful data-science packages and tools.+FROM datarobot/python3-dropin-env-base++# Install the list of core requirements, e.g. sklearn, numpy, pandas, flask.+# **Don't modify this file!**+COPY dr_requirements.txt dr_requirements.txt++# '--upgrade-strategy eager' will upgrade installed dependencies+# according to package requirements or to the latest+RUN pip3 install -U --upgrade-strategy eager --no-cache-dir --prefer-binary -r dr_requirements.txt  && \+    rm -rf dr_requirements.txt++# Install the list of custom Python requirements, e.g. keras, xgboost, etc.+COPY requirements.txt requirements.txt+RUN pip3 install -r requirements.txt --no-cache-dir && \+    rm -rf requirements.txt++# get and install  julia 1.5.4++ENV HOME /opt+RUN apt-get update && apt-get install wget -y && \+    wget https://julialang-s3.julialang.org/bin/linux/x64/1.5/julia-1.5.4-linux-x86_64.tar.gz && \+    tar zxvf julia-1.5.4-linux-x86_64.tar.gz && \+    mkdir $HOME/.julia && \+    mkdir $HOME/julia++ENV JULIA_HOME /julia-1.5.4+ENV JULIA_SYS_IMAGE $HOME/julia/sys.so+ENV JULIA_PROJECT $HOME/julia+ENV PATH $PATH:$JULIA_HOME/bin+COPY sysim.jl $HOME/julia/sysim.jl+COPY . /opt/code+ENV PYTHON /usr/local/bin/python +RUN julia "$HOME/julia/sysim.jl" && \+    chmod -R 777 $HOME/.julia/logs/manifest_usage.toml && \+    rm julia-1.5.4-linux-x86_64.tar.gz++ENV CODE_DIR=/opt/code +ENV ADDRESS=0.0.0.0:8080 WITH_ERROR_SERVER=1++#Uncomment the following line to switch from Flask to uwsgi server+ENV PRODUCTION=1 MAX_WORKERS=1 SHOW_STACKTRACE=1

Maybe comment it away, as we have in other envs.

timsetsfire

comment created time in 20 hours

Pull request review commentdatarobot/datarobot-user-models

[RAPTOR-5203] Add support for Julia Inference Models

+# This is the default base image for use with user models and workflows.+# It contains a variety of common useful data-science packages and tools.+FROM datarobot/python3-dropin-env-base++# Install the list of core requirements, e.g. sklearn, numpy, pandas, flask.+# **Don't modify this file!**+COPY dr_requirements.txt dr_requirements.txt++# '--upgrade-strategy eager' will upgrade installed dependencies+# according to package requirements or to the latest+RUN pip3 install -U --upgrade-strategy eager --no-cache-dir --prefer-binary -r dr_requirements.txt  && \+    rm -rf dr_requirements.txt++# Install the list of custom Python requirements, e.g. keras, xgboost, etc.+COPY requirements.txt requirements.txt+RUN pip3 install -r requirements.txt --no-cache-dir && \+    rm -rf requirements.txt++# get and install  julia 1.5.4++ENV HOME /opt+RUN apt-get update && apt-get install wget -y && \+    wget https://julialang-s3.julialang.org/bin/linux/x64/1.5/julia-1.5.4-linux-x86_64.tar.gz && \+    tar zxvf julia-1.5.4-linux-x86_64.tar.gz && \+    mkdir $HOME/.julia && \+    mkdir $HOME/julia++ENV JULIA_HOME /julia-1.5.4+ENV JULIA_SYS_IMAGE $HOME/julia/sys.so

also ENVs can be combine into a one directive, except for JULIA_HOME as it is used later

timsetsfire

comment created time in 20 hours

Pull request review commentdatarobot/datarobot-user-models

[RAPTOR-5203] Add support for Julia Inference Models

+pyarrow==0.14.1+datarobot-drum==1.4.16

latest drum is 1.5.5

timsetsfire

comment created time in 20 hours

Pull request review commentdatarobot/datarobot-user-models

[RAPTOR-5203] Add support for Julia Inference Models

+# This is the default base image for use with user models and workflows.+# It contains a variety of common useful data-science packages and tools.+FROM datarobot/python3-dropin-env-base++# Install the list of core requirements, e.g. sklearn, numpy, pandas, flask.+# **Don't modify this file!**+COPY dr_requirements.txt dr_requirements.txt

It's recommended to have as little layers as possible:

copy both requirements COPY requirements_drum.txt requirements_dropin.txt ./

install both requirments files (adjust paths) RUN pip install -r /tmp/requirements_drum.txt -r /tmp/requirements_dropin.txt

timsetsfire

comment created time in 21 hours

Pull request review commentdatarobot/datarobot-user-models

[RAPTOR-5203] Add support for Julia Inference Models

+module Custom++using PyCall +export load_model, unstructured_predict++function to_bytes(n::Integer; bigendian=true, len=sizeof(n))+    bytes = Array{UInt8}(undef, len)+    for byte in (bigendian ? (1:len) : reverse(1:len))+        bytes[byte] = n & 0xff+        n >>= 8+    end+    return bytes+end++function load_model(code_dir)  +    return nothing+end++function score_unstructured(data; model = nothing, kwargs = nothing)+    println("Model: $model")+    println("Incoming content type params: $kwargs")+    println("Incoming data type $(typeof(data))")+    println("Incoming data: $data")+    println("Incoming query params: $(kwargs.query)")+    +    if haskey(kwargs.query, "ret_mode")+        ret_mode = kwargs.query["ret_mode"]+    else +        ret_mode = nothing+    end++    word_count = split( replace( data, "\n" => " "), " ")+    word_count = length(word_count)++    if ret_mode == "binary"  +        ret_data = to_bytes(word_count)+        ret_kwargs = PyCall.PyDict( Dict("mimetype" => "application/octet-stream"))+        ret = PyCall.pybytes(ret_data), ret_kwargs+    else +        ret_kwargs = PyCall.PyDict( Dict("mimetype" => kwargs.mimetype)) +        ret = string(word_count)+        #, ret_kwargs+    end+    println("outgoing")+    println(ret)+    return ret+end++end++# print("Model: ", model)

Please remove commented code (if it doesn't bring any value)

timsetsfire

comment created time in 21 hours

Pull request review commentdatarobot/datarobot-user-models

[RAPTOR-5203] Add support for Julia Inference Models

+numpy>=1.16.0,<1.19.0+pandas==1.1.0

Could this env use pandas==1.0.5 as higher versions have a memory leak.

timsetsfire

comment created time in 20 hours

Pull request review commentdatarobot/datarobot-user-models

[RAPTOR-5203] Add support for Julia Inference Models

+import logging+import numpy+import os+import pandas as pd+import time++from datarobot_drum.drum.common import (+    LOGGER_NAME_PREFIX,+    TargetType,+    CustomHooks,+    REGRESSION_PRED_COLUMN,+    UnstructuredDtoKeys,+    PayloadFormat,+    SupportedPayloadFormats,+    StructuredDtoKeys,+)+from datarobot_drum.drum.exceptions import DrumCommonException+from datarobot_drum.drum.language_predictors.base_language_predictor import BaseLanguagePredictor++logger = logging.getLogger(LOGGER_NAME_PREFIX + "." + __name__)++CUR_DIR = os.path.dirname(os.path.abspath(__file__))+JL_SCORE_PATH = os.path.join(CUR_DIR, "score.jl")+JL_SYSIMAGE_PATH = os.environ.get("JULIA_SYS_IMAGE")+JL_PROJECT = os.environ.get("JULIA_PROJECT", CUR_DIR)+JL_COMMON_PATH = os.path.abspath(os.path.join(CUR_DIR, "..", "julia_common_code", "common.jl",))+JL_INIT = ["--history-file=no"]+logger.info(f"Julia project director set as {JL_PROJECT}")++try:+    from julia.api import Julia+    from julia.api import LibJulia, JuliaInfo+except ImportError:+    error_message = (+        "julia package is not installed."+        "Install julia using 'pip install julia==0.5.6'"+        "Available for Python>=3.4 and Python <= 3.8"+    )+    logger.error(error_message)+    raise DrumCommonException(error_message)++## need better invocation here+try:+    jl = Julia(sysimage=JL_SYSIMAGE_PATH, init_julia=JL_INIT)+except Exception as error_message:+    logger.error(error_message)+    jl = Julia(init_julia=JL_INIT)+logger.info("Julia ready!")+from julia import Base++logger.info(f"julia was started with {Base.julia_cmd()}")+jl.eval(f'using Pkg; Pkg.activate("{JL_PROJECT}"); Pkg.instantiate()')+++class JlPredictor(BaseLanguagePredictor):+    def __init__(self,):+        super(JlPredictor, self).__init__()++    def configure(self, params):+        super(JlPredictor, self).configure(params)+        logger.info(f"loading {JL_SCORE_PATH}")+        jl.eval(f'include("{JL_SCORE_PATH}")')+        logger.info(f"{JL_SCORE_PATH} loaded")+        from julia import Main++        Main.init(self._code_dir, self._target_type.value)+        self._model = Main.load_serialized_model(self._code_dir)++    @property+    def supported_payload_formats(self):+        formats = SupportedPayloadFormats()+        formats.add(PayloadFormat.CSV)+        # formats.add(PayloadFormat.MTX)+        return formats++    def has_read_input_data_hook(self):+        return Main.defined_hooks["read_input_data"]++    def predict(self, **kwargs):+        from julia import Main++        input_binary_data = kwargs.get(StructuredDtoKeys.BINARY_DATA)+        mimetype = kwargs.get(StructuredDtoKeys.MIMETYPE)+        start_predict = time.time()

Remove start_predict, end_predict, execution_time, self.monitor if your PR goes in after mine: https://github.com/datarobot/datarobot-user-models/pull/326

timsetsfire

comment created time in 21 hours

Pull request review commentdatarobot/datarobot-user-models

[RAPTOR-5203] Add support for Julia Inference Models

+## Flux Model ++Source the appropriate Julia Environment++`export JULIA_PROJECT=/Users/timothy.whittaker/Desktop/mlops-experiments/python-julia-sys-image`++`julia --project=$JULIA_PROJECT`++Install `Flux` and `BSON` in your Julia environment if not already available++Again, this is a slow startup time.  For scoring and serving it is fine, but it will always take some time to start everything up, therefore the recommendation is to use a system image of your environment to speed up things considerably.  ++## Scoring++drum score --code-dir model_templates/inference/julia/jl_custom --target-type regression --input tests/testdata/boston_housing_inference.csv --verbose --logging-level info++## Serving with Docker++drum server --code-dir model_templates/inference/julia/jl_custom --target-type regression --input tests/testdata/boston_housing_inference.csv --verbose --logging-level info++++++

Please remove these extra spaces :)

timsetsfire

comment created time in 21 hours

Pull request review commentdatarobot/datarobot-user-models

[RAPTOR-5203] Add support for Julia Inference Models

 Include any necessary hooks in a file called `custom.py` for Python models or `c Type checking methods can be used to verify types: - in Python`isinstance(data, str)` or `isinstance(data, bytes)` - in R `is.character(data)` or `is.raw(data)`+- in Julia `data isa String` or `data is Base.CodeUnits`

typo is a

timsetsfire

comment created time in 21 hours

push eventdatarobot/datarobot-user-models

Scott Powers

commit sha 4099844a979fe038c9d3b66dd497f8f61cfeb511

[RAPTOR-5194 RAPTOR-5195]fit time validation (#325) * Adds validation of typeschema yaml format. * Validation schema documentation. * test datatype schema * test schema * incorrect import * perform schema validation at during fit or push * add tests * black * pr comments * use logger * minor fixes * share loading of input file

view details

Yakov Goldberg

commit sha 2aa988c4c2d2270263fe15c4a3f995c5e06eb532

[RAPTOR-5277] validate classification probs add up to one for all the predictors

view details

Yakov Goldberg

commit sha 30ad7712b81f520b4249d23d3369cbc45cbedfd5

Code clean up and optimization

view details

Yakov Goldberg

commit sha 5fd333498e7d5dc6622c2a2a74b7da4e0783193f

Move probs add up into validation. Refactor validation

view details

Yakov Goldberg

commit sha 6dd1a89180290811f701517a6880ee444fdedb44

typos and drum->DRUM capitalization

view details

push time in a day

push eventdatarobot/datarobot-user-models

Yakov Goldberg

commit sha 367829efc7f6383e747ac1cbc1b3aadc21c77c55

typos and drum->DRUM capitalization

view details

push time in a day