- PVSM.RU - https://www.pvsm.ru -
ML.NET [1] is an open-source and cross-platform machine learning framework (Windows, Linux, macOS) for .NET developers. Using ML.NET [1], developers can leverage their existing tools and skillsets to develop and infuse custom AI into their applications by creating custom machine learning models for common scenarios like Sentiment Analysis, Recommendation, Image Classification and more!.
Today we’re announcing the ML.NET [1] 1.0 RC (Release Candidate) (version 1.0.0-preview
) which is the last preview release before releasing the final ML.NET [1] 1.0 RTM in 2019 Q2 calendar year.
Soon we will be ending the first main milestone of a great journey in the open that started on May 2018 when releasing ML.NET 0.1 as open source. Since then we’ve been releasing monthly, 12 preview releases so far, as shown in the roadmap below:
In this release (ML.NET [1] 1.0 RC) we have initially concluded our main API changes. For the next sprint we are focusing on improving documentation and samples and addressing major critical issues if needed.
The goal is to avoid any new breaking changes moving forward.
Segregation of stable vs. preview version of ML.NET packages: Heading ML.NET 1.0, most of the functionality in ML.NET (around 95%) is going to be released as stable (version 1.0).
You can review the reference list of the ‘stable’ packages and classes here [2].
However, there are a few feature-areas which still won’t be in RTM state when releasing ML.NET 1.0. Those features still kept as preview are being categorized as preview packages with the version 0.12.0-preview
.
The main packages that will continue in preview state after ML.NET 1.0 is released are the following (0.12 version packages
):
You can review the full reference list of “after 1.0” preview packages and classes (0.12.0-preview) here [3].
IDataView moved to Microsoft.ML namespace : One change in this release is that we have moved IDataView back into Microsoft.ML namespace based on feedback that we received.
TensorFlow-support fixes: TensorFlow is an open source machine learning framework used for deep learning scenarios (such as computer vision and natural language processing). ML.NET [1] has support for using TensorFlow models, but in ML.NET version 0.11 there were a few issues that have been fixed for the 1.0 RC release.
You can review an example of ML.NET [1] code running a TensorFlow model here [4].
Release Notes for ML.NET 1.0 RC: You can check out additional release notes for 1.0 RC here [5].
For your convenience, if you are moving your code from ML.NET [1] v0.11 to v0.12, you can check out the breaking changes list [6] that impacted our samples.
If you are using ML.NET [1] in your app and looking to go into production, you can talk to an engineer on the ML.NET [1] team to:
Fill out this form [7] and leave your contact information at the end if you’d like someone from the ML.NET [1] team to contact you.
As mentioned, ML.NET 1.0 is almost here! You can get ready before it releases by researching the following resources:
Get started with ML.NET here [8].
Next, going further explore some other resources:
We will appreciate your feedback by filing issues with any suggestions or enhancements in the ML.NET GitHub repo [13] to help us shape ML.NET [1] and make .NET a great platform of choice for Machine Learning.
Thanks and happy coding with ML.NET [1]!
The ML.NET [1] Team.
This blog was authored by Cesar de la Torre plus additional contributions of the ML.NET [1] team
Principal Program Manager, .NET
Автор: msgeek
Источник [15]
Сайт-источник PVSM.RU: https://www.pvsm.ru
Путь до страницы источника: https://www.pvsm.ru/programmirovanie/314593
Ссылки в тексте:
[1] ML.NET: https://dot.net/ml
[2] reference list of the ‘stable’ packages and classes here: https://docs.microsoft.com/dotnet/api/index?view=ml-dotnet
[3] reference list of “after 1.0” preview packages and classes (0.12.0-preview) here: https://docs.microsoft.com/dotnet/api/microsoft.ml?view=ml-dotnet-preview
[4] here: https://github.com/dotnet/machinelearning-samples/tree/master/samples/csharp/getting-started/DeepLearning_ImageClassification_TensorFlow
[5] here: https://github.com/dotnet/machinelearning/blob/master/docs/release-notes/1.0.0-preview/release-1.0.0-preview.md
[6] breaking changes list: https://github.com/dotnet/machinelearning-samples/blob/master/docs/migrations/breaking-changes-1.0.0-preview.md
[7] this form: https://www.research.net/r/mlnet-in-production
[8] ML.NET here: https://www.microsoft.com/net/learn/apps/machine-learning-and-ai/ml-dotnet/get-started
[9] Microsoft Docs ML.NET Guide: https://docs.microsoft.com/dotnet/machine-learning/
[10] machinelearning-samples GitHub repo: https://github.com/dotnet/machinelearning-samples
[11] here: https://docs.microsoft.com/dotnet/machine-learning/basic-concepts-model-training-in-mldotnet
[12] here: https://docs.microsoft.com/dotnet/machine-learning/how-to-guides/
[13] ML.NET GitHub repo: https://github.com/dotnet/machinelearning
[14] Cesar De la Torre: https://devblogs.microsoft.com/dotnet/author/cesardl/
[15] Источник: https://habr.com/ru/post/447936/?utm_campaign=447936
Нажмите здесь для печати.