v0.4.4

What's Changed

• Docs: Use modified_model in Composites example by @rodrigobdz in #64
• Readme: Fix typo 'exerpt' by @rodrigobdz in #65
• Core: Fix typo 'existance' by @rodrigobdz in #66
• Composites: Fix typo 'occurence' by @rodrigobdz in #67
• Readme: Correct order of comments describing vars by @rodrigobdz in #68
• Fixtures for linear modules, test MergeBatchNorm by @chr5tphr in #69
• Added new abstract types concerning Convolution by @chr5tphr in #70
• Fix MergeBatchNorm for all Convolutions by @chr5tphr in #71
• Docs: Clarification about requirements Modules/ Canonizers by @chr5tphr in #72
• Fixed flipped seismic cmap by @chr5tphr in #73
• Example: Fix typo 'orignal' by @rodrigobdz in #74
• Attribution: Fix typos 'convien[i]ence' by @rodrigobdz in #75
• How-to for Rules, Composites and Canonizers by @chr5tphr in #78
• Fixed code examples for How-To/Composites by @chr5tphr in #79
• Fix: Occlusion stride resolve by @chr5tphr in #80
• Assume batch-dimension in window for occlusion by @chr5tphr in #81
• Docs: Added How-To for using Attributors by @chr5tphr in #82
• Norm-mode for imgify and imsave by @chr5tphr in #83
• Added grid function to imgify by @chr5tphr in #84
• Compute norm bounds instead of directly applying by @chr5tphr in #85
• Remove norm mode 'absolute', make symmetric flag by @chr5tphr in #86
• Allow single-digit index for CMSL by @chr5tphr in #87
• Cast vmin and vmax to ndarray in imgify by @chr5tphr in #88
• How-To for Visualization of Results by @chr5tphr in #89

New Contributors

• @rodrigobdz made their first contribution in #64

Full Changelog: 0.4.3...0.4.4