<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0">
  <channel>
    <title>PyPI recent updates for stream-dataset</title>
    <link>https://pypi.org/project/stream-dataset/</link>
    <description>Recent updates to the Python Package Index for stream-dataset</description>
    <language>en</language>    <item>
      <title>0.1.6</title>
      <link>https://pypi.org/project/stream-dataset/0.1.6/</link>
      <description>STREAM (Sequential Tasks Review to Evaluate Artificial Memory) is a dataset of 12 diverse sequential tasks to assess neural networks’ memory. Scalable in complexity and sequence length, it covers pattern completion, copy tasks, forecasting, bracket matching, and sorting—ideal for comparing architectures on memory retention and sequential reasoning.</description>
<author>yannis.bendiouis@gmail.com</author>      <pubDate>Sun, 03 May 2026 17:43:44 GMT</pubDate>
    </item>    <item>
      <title>0.1.5</title>
      <link>https://pypi.org/project/stream-dataset/0.1.5/</link>
      <description>STREAM (Sequential Tasks Review to Evaluate Artificial Memory) is a dataset of 12 diverse sequential tasks to assess neural networks’ memory. Scalable in complexity and sequence length, it covers pattern completion, copy tasks, forecasting, bracket matching, and sorting—ideal for comparing architectures on memory retention and sequential reasoning.</description>
<author>yannis.bendiouis@gmail.com</author>      <pubDate>Sat, 04 Apr 2026 17:30:41 GMT</pubDate>
    </item>    <item>
      <title>0.1.4</title>
      <link>https://pypi.org/project/stream-dataset/0.1.4/</link>
      <description>STREAM (Sequential Tasks Review to Evaluate Artificial Memory) is a dataset of 12 diverse sequential tasks to assess neural networks’ memory. Scalable in complexity and sequence length, it covers pattern completion, copy tasks, forecasting, bracket matching, and sorting—ideal for comparing architectures on memory retention and sequential reasoning.</description>
<author>yannis.bendiouis@gmail.com</author>      <pubDate>Sat, 04 Apr 2026 17:16:55 GMT</pubDate>
    </item>    <item>
      <title>0.1.3</title>
      <link>https://pypi.org/project/stream-dataset/0.1.3/</link>
      <description>STREAM (Sequential Tasks Review to Evaluate Artificial Memory) is a dataset of 12 diverse sequential tasks to assess neural networks’ memory. Scalable in complexity and sequence length, it covers pattern completion, copy tasks, forecasting, bracket matching, and sorting—ideal for comparing architectures on memory retention and sequential reasoning.</description>
<author>yannis.bendiouis@gmail.com</author>      <pubDate>Sat, 04 Apr 2026 16:24:53 GMT</pubDate>
    </item>    <item>
      <title>0.1.2</title>
      <link>https://pypi.org/project/stream-dataset/0.1.2/</link>
      <description>STREAM (Sequential Tasks Review to Evaluate Artificial Memory) is a dataset of 12 diverse sequential tasks to assess neural networks’ memory. Scalable in complexity and sequence length, it covers pattern completion, copy tasks, forecasting, bracket matching, and sorting—ideal for comparing architectures on memory retention and sequential reasoning.</description>
<author>yannis.bendiouis@gmail.com</author>      <pubDate>Fri, 03 Apr 2026 18:26:51 GMT</pubDate>
    </item>    <item>
      <title>0.1.1</title>
      <link>https://pypi.org/project/stream-dataset/0.1.1/</link>
      <description>STREAM (Sequential Tasks Review to Evaluate Artificial Memory) is a dataset of 12 diverse sequential tasks to assess neural networks’ memory. Scalable in complexity and sequence length, it covers pattern completion, copy tasks, forecasting, bracket matching, and sorting—ideal for comparing architectures on memory retention and sequential reasoning.</description>
<author>yannis.bendiouis@gmail.com</author>      <pubDate>Fri, 03 Apr 2026 12:13:33 GMT</pubDate>
    </item>    <item>
      <title>0.1.0</title>
      <link>https://pypi.org/project/stream-dataset/0.1.0/</link>
      <description>STREAM (Sequential Tasks Review to Evaluate Artificial Memory) is a dataset of 12 diverse sequential tasks to assess neural networks’ memory. Scalable in complexity and sequence length, it covers pattern completion, copy tasks, forecasting, bracket matching, and sorting—ideal for comparing architectures on memory retention and sequential reasoning.</description>
<author>yannis.bendiouis@gmail.com</author>      <pubDate>Fri, 03 Apr 2026 12:06:33 GMT</pubDate>
    </item>  </channel>
</rss>