search

hybridlca / pyspa

A python package for conducting structural path analysis on square technological matrices of process or input-output data, using environmental, social and/or financial satellites
GNU General Public License v3.0
25 stars 4 forks source link

Question about infosheet #5

Open urnotsunn opened 1 year ago

urnotsunn commented 1 year ago

halo and thanks for the excellent pyspa. i have a problem with this code. i have 18 environment flows and I want to calculate them together. based on your documents, I have set them into the infosheet together in one Excel, as shown in the following. but the results of "pyspa" is wrong. it can't select any pathways. could you please tell me why? d347930e1bd665b34266ddfba78dfb5 aaa0f1fa496c3181826b14f4e935cbc

andrestephan1 commented 1 year ago

Hi @urnotsunn ,

Thanks for your interest in pyspa. Could you please give more information? What's the size of your A matrix? How are you calling the code? You seem to have some direct and total coefficients that are both 0 for given sectors/processes. What is the response of the code to your call? With these, I might get an idea of what is happening. Know that we've successfully used pyspa on multiple indicators and on matrices up to 100k by 100k.

André

urnotsunn commented 1 year ago

Hi André,

Thanks for your reply. The dimension of A matrix is 19565*19565, which is derived from ecoinvent v3.8. I calculate the direct coefficient (E) by multiplying C (characterization factors) and S (satellite intensity). I don't know why some direct and total coefficients are both 0. I have 18 environment flows and I want to calculate them together. Based on your documents, I have set them into the "infosheet" together in one Excel. But the results of "pyspa" is wrong. it only selects one pathway. In theory, my study should have been able to extract many pathways, but the results showed that I could not extract these pathways. Is that because the direct coefficient and the complete coefficient are zero? QQ图片20230924225523 {DH0 $) D$@JXP4 GIF`MUL Thanks again. Looking forward to your reply.

Best regards, Xiaotian Sun