Source: pandas Followup-For: Bug #983328 scipy have rectified the API problem they introduced in 1.6.1, with the fix released in scipy 1.6.2. I've uploaded that to experimental.
The current pandas now passes tests again with scipy 1.6.2. I suggest closing this bug with the future upload of pandas 1.2.3 (or later), which contains pandas PR#40020 (commit b14ffd8) fixing pandas for scipy 1.6.1 by using an explicit dtype for the COO sparse matrices (which is probably a good idea anyway, even though scipy 1.6.2 fixes the unexpected change in behaviour). -- System Information: Debian Release: bullseye/sid APT prefers testing-security APT policy: (500, 'testing-security'), (500, 'unstable'), (1, 'experimental') Architecture: amd64 (x86_64) Foreign Architectures: i386 Kernel: Linux 5.10.0-4-amd64 (SMP w/8 CPU threads) Kernel taint flags: TAINT_PROPRIETARY_MODULE, TAINT_WARN, TAINT_OOT_MODULE Locale: LANG=en_AU.UTF-8, LC_CTYPE=en_AU.UTF-8 (charmap=UTF-8), LANGUAGE=en_AU:en Shell: /bin/sh linked to /usr/bin/dash Init: systemd (via /run/systemd/system) LSM: AppArmor: enabled
