aws/sagemaker-python-sdk issues and pull requests

#1814 - Incorrect callable function signatures

Issue - State: closed - Opened by ehsanmok about 5 years ago - 2 comments
Labels: type: documentation, contributions welcome

#1802 - breaking: merge v2 changes into master

Pull Request - State: closed - Opened by laurenyu about 5 years ago - 8 comments

#1784 - Make cli handling more useful

Issue - State: open - Opened by ehsanmok about 5 years ago - 2 comments
Labels: type: feature request

#1779 - Content gap: Documentation for the Built-in algorithms is incomplete

Issue - State: closed - Opened by mchoi8739 about 5 years ago - 1 comment
Labels: type: documentation

#1703 - Model Monitor on Batch Tansform

Issue - State: open - Opened by Chichilele about 5 years ago - 1 comment
Labels: type: feature request, component:model monitor

#1687 - Formatting error on PR

Issue - State: open - Opened by yijiezh about 5 years ago - 4 comments
Labels: bug, Pending information

#1657 - Use standard python logging facilities instead of printing to stdout

Issue - State: closed - Opened by saulbein over 5 years ago - 5 comments
Labels: type: feature request, contributions welcome

#1649 - input_fn gets called on inference outputs during inference job

Issue - State: closed - Opened by dshen109 over 5 years ago - 3 comments
Labels: bug, Pending information

#1643 - get_image_uri should return latest version of XGBoost by default instead of a previous version

Issue - State: closed - Opened by asimjalis over 5 years ago - 4 comments
Labels: type: feature request, Pending information

#1624 - ModelMonitor class doesn't cleanout monitor_schedule_name if create_monitor_schedule() fails.

Issue - State: closed - Opened by jmgray24 over 5 years ago - 5 comments
Labels: bug, component:model monitor

#1623 - Can't deploy model to t3 instance types

Issue - State: closed - Opened by thomelane over 5 years ago - 3 comments
Labels: type: question

#1619 - ScriptProcessor unable to handle millions of output files to be saved to S3

Issue - State: open - Opened by maslick over 5 years ago - 2 comments
Labels: bug, Pending information, component: processing

#1606 - Apply same linting and code style used in the package to the code examples in the docs

Issue - State: open - Opened by metrizable over 5 years ago
Labels: type: documentation, contributions welcome, type: maintenance, component: pysdk-team

#1597 - "local code" strips the entry point path

Issue - State: open - Opened by litaws over 5 years ago - 2 comments
Labels: type: bug, component: local mode, component: pysdk-team

#1591 - More console log info model.deploy()?

Issue - State: open - Opened by hunkim over 5 years ago - 3 comments
Labels: type: feature request, type: logging/error reporting

#1527 - TF 2.2.0 image is not available in eu-west-1

Issue - State: open - Opened by yegortokmakov over 5 years ago - 8 comments
Labels: type: documentation

#1503 - Support `experiment_config` in HyperparameterTuner for unified lineage across training and tuning phases

Issue - State: open - Opened by cfregly over 5 years ago - 6 comments
Labels: type: feature request, component: experiments

#1484 - .pd model deployment error, no SavedModel bundles found!

Issue - State: closed - Opened by Xixiong-Guo over 5 years ago - 3 comments

#1478 - Create a script to aid with migrating from v1.x to v2.0

Issue - State: closed - Opened by laurenyu over 5 years ago - 10 comments
Labels: type: feature request, type: maintenance

#1474 - Minimize **kwargs usage

Issue - State: open - Opened by laurenyu over 5 years ago
Labels: type: maintenance

#1467 - Allowing to instantiate the Framework estimator class to support custom Framework containers

Issue - State: open - Opened by giuseppeporcelli over 5 years ago - 3 comments
Labels: type: feature request, Custom Framework

#1466 - Clean up utils and fw_utils

Issue - State: closed - Opened by laurenyu over 5 years ago - 1 comment
Labels: type: maintenance

#1463 - Refactor sagemaker.session.Session

Issue - State: closed - Opened by laurenyu over 5 years ago - 1 comment
Labels: type: maintenance

#1444 - Clearer interaction of TensorFlowModel with new framework versions

Issue - State: open - Opened by athewsey over 5 years ago - 1 comment
Labels: type: feature request, Tensorflow

#1438 - (TF) estimator.attach() doesn't override latest_training_job for deployment

Issue - State: open - Opened by athewsey over 5 years ago - 2 comments
Labels: type: feature request, Tensorflow, component: hosting

#1429 - Can't override SKLearnModel properties

Issue - State: closed - Opened by thomelane over 5 years ago - 2 comments
Labels: type: feature request, scikit-learn, component: hosting

#1427 - Can't pass new `entry_point` to `estimator.deploy()`

Issue - State: closed - Opened by thomelane over 5 years ago - 2 comments
Labels: type: feature request, Pending information, scikit-learn

#1400 - [Request] SKLearnProcessor no image_uri support

Issue - State: closed - Opened by bmaingret over 5 years ago - 3 comments
Labels: type: feature request

#1389 - Elastic Inference for Pytorch 1.4.0

Issue - State: closed - Opened by raguiar2 over 5 years ago - 4 comments
Labels: type: feature request, PyTorch

#1354 - Local deploy failure for MXNet estimator due to non-uniform output uri

Issue - State: closed - Opened by ehsanmok over 5 years ago - 11 comments
Labels: bug, Local Mode

#1324 - Different `entry_point` for different models in Multi-model

Issue - State: open - Opened by dshahrokhian over 5 years ago - 1 comment
Labels: type: feature request, component: hosting

#1318 - The link is not working

Issue - State: closed - Opened by pingsutw over 5 years ago - 3 comments
Labels: type: documentation

#1297 - Local deployment is not working on Windows 10

Issue - State: open - Opened by ViktorStepanukCN over 5 years ago - 5 comments
Labels: type: feature request, component: Inference APIs and Interfaces, OS: Windows

#1282 - ScriptProcessor does not allow us to specify S3 SSE-KMS

Issue - State: closed - Opened by jerrygb over 5 years ago - 4 comments
Labels: type: feature request, component: processing

#1260 - SM_CURRENT_HOST changes for spot instance training

Issue - State: closed - Opened by sivakhno over 5 years ago - 1 comment
Labels: type: documentation, XGBoost, component: training

#1248 - Please add support for `requirements.txt` in ScriptProcessor similar to other "Script Mode" parts of the SageMaker Python SDK

Issue - State: open - Opened by cfregly over 5 years ago - 19 comments
Labels: type: feature request, component: processing

#1200 - Unable to use this library in AWS Lambda due to package size exceeded max limit

Issue - State: open - Opened by nemalipuri almost 6 years ago - 26 comments
Labels: type: feature request, type: documentation, Lambda

#1160 - Cannot modify SM_MODULE_DIR so cannot control output location of source.tar.gz on s3

Issue - State: closed - Opened by drewthayer almost 6 years ago - 2 comments
Labels: type: feature request

#1140 - Local mode crashes on Windows when running locally

Issue - State: open - Opened by Serhiy-Shekhovtsov almost 6 years ago - 1 comment
Labels: type: bug, contributions welcome, component: local mode, OS: Windows

#1136 - Numpy de/serialization is buggy for XGBoost endpoint

Issue - State: closed - Opened by tigerhawkvok almost 6 years ago - 4 comments
Labels: bug, type: documentation, Pending information

#1135 - csv_weights cannot be set

Issue - State: open - Opened by tigerhawkvok almost 6 years ago - 2 comments
Labels: type: bug, type: documentation, XGBoost, component: training

#1127 - GPU utilization is always zero when doing inference

Issue - State: closed - Opened by Andreachen0707 almost 6 years ago - 10 comments

#1124 - No KMS Key support

Issue - State: closed - Opened by samlovestech almost 6 years ago - 3 comments
Labels: bug, contributions welcome

#1123 - Question: updating endpoint with scaling policy

Issue - State: closed - Opened by anotinelg almost 6 years ago - 3 comments
Labels: type: feature request, component: Utility APIs

#1122 - Question: is is possible to specify the variant_name when we deploy a Model

Issue - State: closed - Opened by anotinelg almost 6 years ago - 3 comments
Labels: type: feature request, contributions welcome

#1119 - Increasing the timeout for InvokeEndpoint

Issue - State: open - Opened by velociraptor111 almost 6 years ago - 36 comments
Labels: type: feature request, component: hosting, Good First Issue

#1103 - The method _run_clone_command does not support git/CodeCommit ssh URLs

Issue - State: closed - Opened by jasm37 almost 6 years ago - 2 comments
Labels: bug, contributions welcome

#1092 - SageMaker Airflow Operators should handle sessions

Issue - State: closed - Opened by andrewcstewart almost 6 years ago - 3 comments
Labels: type: feature request, contributions welcome

#1090 - Add Tensorboard API to Pytorch

Issue - State: closed - Opened by tanguycdls almost 6 years ago - 3 comments
Labels: type: feature request, contributions welcome

#1088 - Edit neo docs to include currently supported target family

Issue - State: closed - Opened by w601sxs almost 6 years ago - 1 comment
Labels: type: documentation

#1079 - np.int is int64 on many platforms, so why is np.int here ascribed int32 type?

Issue - State: closed - Opened by jharrang almost 6 years ago - 3 comments
Labels: type: feature request, contributions welcome

#1065 - Failed to import code copied into the /opt/ml/model/code/lib directory

Issue - State: closed - Opened by anotinelg about 6 years ago - 2 comments
Labels: bug, type: feature request, MXNet

#1057 - Internal server error when using OpenNMT-tf

Issue - State: closed - Opened by anasir-ureed about 6 years ago - 19 comments
Labels: bug, type: question

#1023 - write_sparse_tensor_to_spmatrix

Issue - State: closed - Opened by Jomonsugi about 6 years ago - 6 comments
Labels: type: feature request, component: Inference APIs

#993 - RealTimePredictor.accept values not correctly checked

Issue - State: closed - Opened by vanoudh about 6 years ago - 6 comments
Labels: type: feature request, type: documentation

#984 - Limited Size of Hyperparameters argument for Estimator

Issue - State: closed - Opened by InzamamAnwar about 6 years ago - 7 comments
Labels: type: feature request, component: training

#975 - Missing dependencies in SKLearn based endpoint/Batch Transform job

Issue - State: closed - Opened by lihip about 6 years ago - 9 comments
Labels: bug, type: feature request

#937 - PyTorch: increasing --shm-size to allow multiprocessing data loaders

Issue - State: closed - Opened by ksanjeevan about 6 years ago - 23 comments
Labels: type: feature request

#929 - Local Mode Endpoints can't be created from hosted training jobs

Issue - State: closed - Opened by rabowskyb about 6 years ago - 9 comments
Labels: contributions welcome, component: local mode

#868 - `requirements.txt` not working for PytorchModel

Issue - State: closed - Opened by dshahrokhian over 6 years ago - 21 comments

#846 - "No such file or directory: '/opt/ml/model/model.joblib'" occured at deploy in the local mode of windows

Issue - State: closed - Opened by xnaiman over 6 years ago - 2 comments
Labels: type: feature request, Local Mode, OS: Windows

#845 - I want env parameter for fit method of estimators

Issue - State: closed - Opened by xnaiman over 6 years ago - 4 comments
Labels: type: feature request

#844 - Invalid specification of input data in windows local mode

Issue - State: closed - Opened by xnaiman over 6 years ago - 2 comments
Labels: type: feature request, Local Mode, OS: Windows

#826 - In Local mode, Batch Transform ManifestFile option gives the endpoint the ManifestFile as input, which seems incorrect

Issue - State: closed - Opened by zmjjmz over 6 years ago - 4 comments
Labels: type: feature request, contributions welcome, Local Mode

GitHub / aws/sagemaker-python-sdk issues and pull requests