So I’m at a point now where I’m working up a bit of code that will be reused with a different set of fit
s. Basically a processing chain of fluid.robustscale~
→ fluid.pca~
→ fluid.umap~
→ fluid.normalize~
.
Other than having to zl slice 2
after fluid.pca~
(as per this thread) setting up that processing chain is fairly painless.
However, if I run it once and get the fit
s that I want, and then want to apply them to a different fluid.dataset~
(to prep for some regression, where I want the dimensions scaled/processed the same on each side) the code no longer works as fit
only passes on a bang
and not the name of the internally generated dataset.
I know that in this thread I was arguing for fluid.pca~
's fit
message to output it’s variance, but perhaps that isn’t ideal. I’ll bump that thread with some thoughts.
But for this, it means a completely different set of code if you are fittransform
-ing vs just fit
-ing. So perhaps fit
should also output its own name (if internally generated) to keep parity with fittransofrm
in terms of coding workflow. Don’t know offhand if there are other messages which daisychain in a similar way, but those should perhaps do the same?
----------begin_max5_patcher----------
1511.3oc2Z0sbiZCE9ZmmBMLSm1NiWJ5GDPupW19LzoSFYibB6hAJHRxt6r4
YuRBvFuACxF4kLMLwYPQvQeme+NR9q2sxYS9K7JGvuC9avpUe8tUqzCoFXU6
8qb1ydYaJqROMms462yyDNqa9eB9KhlwYY+r.THmFXOuph8.uBjmAd9QdFnt
JI6A4coeFrK4villjw2lWmoedT6fEk7J4qmIRxyte3YvDaeT99tujuUzrxiP
DW+0.RnmqW.53kWvZ.zWN3Z.F65A9m1WQRrdImu4iev2ua4jUuOIKkKzvDdb
v7ZQ2ndpA+1c2o9X8L0WhGSp.w47pLA347xO4LB7hHZHPCOAdXzAzg7NC5nK
D5vtfr7mA+E3YlDehb4ck6YoIegCXf3jc6.wLAqhKZ8MDOxAUr87Y3ePNuBL
LJzEIcPfP+ScPhjpP4GJUHg1SEtKOSrisk2WY0WsFLpZU8zURnpW2J6yfZ6d
SNShb8K+O4oOwEIaYNWgsHi+rbw8FSwtz5jX2hsrWclNBxG8FWLZiGFpm5oA
FhOWvadZGm0xeGzAL7Bb.QWAnay07VTezOZH3FFofKFOPDENx.7NHViFDqnA
wJzdFXYHQAOKFLAjarvHRvaCAn9WIjodWf40hPtT9Z4.cXyzH1KRF7SNb4io
RDGMCmZJ7BLzHaGIeHSpAwynvfSwNkrFDgmC1QKT.8CbwD16PYMdURdz.d39
WsGNdYBpar0sEIeEH3UBjAXmRcw8fd.QkQCNmL3TxM1faGE.JHTWEWk8tO9w
j4p.7u0JfywgRwklefMMfklqHJkCpp2Tw+2Z4i.jqOoFnx9Lp88uZF0DzRxn
VQktB7PImMVwPJ5ZYTSfKD5Pt83P2Qd9WdJgop6KJYYU6j+6e8F3JHMzMjmk
4VnGuHdJxyPxkRdFG9+MxycAK1l7LNX4HOeviBTiPAddgTDbr.J3MfPMlt3D
pOnFl17OHyZUPyUBdx6Xp0cP15Tqw328Tq6vt0oViguaoVSB7rO0ZTz6Fp0F
.8AYV6MG6MhtPDKgtfsJBR58hqkIwtjzTdL34Dwi.YNu378.4BXCubBxkCz5
IUSW.2nZZJ.dI7CnuGoGbtPmhxjrw7dT63nhZfmmKzKp6R52H4QiQWYfCYgp
K9AvXT.PPnlBfDPvviWHBQsGq5TliBUkhbvzhnkAsBPBXyz3E446hh5YagPY
MPjY3UlZXCK6ggws2xz3cAm+oWAUI6KR4E4i6d2YyoPW3wRgDJApRNZ.0mco
4rgM6vgM63asYu9KIfvwhniNe8.hAr80160G+64b6gWxlLhsG9+MogSlsZLi
daE+PYeg8izIApdfliMObYhzaq2ovsAvlPNg2GQ5LDX.k2yYmCtwXdSsPjmM
VqaMcvSZ.W2mS3.OXBqniRoTV0UvKummw1jx6WS9Ga2LSkDGRCTHVQrs+daf
CjEsHVHINZoN+qsobV4zV8A4kbsD5CfKS76l5c63kmTyB7GUr8EUilIuUEn1
33dV9Prp1U.zz7XiY8gKTI7JMszc.Vbbi13mhMrfda7vf9Evffq8jQWlsxPU
JWkT2C3MoaPXnKwq+OTUVAj8pmSVlx4WVytcl+AOEkY0r60t2F52kt2yu6ao
jd0qF+TcRUdc41tUS2QTCNthikJfjL8Nf2eRnlIMnd2XIAMQRd1PRXSjDwFR
R8RfSIInMjDx.Lo9JVza4jWFyKO+AsXWQGdxjNHZ3rDcfI9L9jgE87Ps+hIY
rIfVc3G1Weq1ywIEsZi4ls+r9kLUjCFaAIYjOD7VnM8WJASnlK4e.tpzaQVI
LwDQaEPZR4DrMJmn1TC.ZJIYiRjpFulL7CBskjlLQpMLSlvjAirkjlR4EXCA
YRkWnUfjII9sgKNw.O7NWlYKnorQDaTIKv.MmM.Tno9BvSDTSGBrhhm3kUsy
VKCY+QeLWmHNbs91jrla06grSI+ojt4q6CwgUJ6GRHaFptr4Hido8K+ly9bY
J8r5j1JWRzIEot2K0wKUUzdZV5Vzt6a28ePjzqAI
-----------end_max5_patcher-----------