我知道这个问题措辞很糟糕，但我想不出如何更好地表达它。
我有两个数据帧，一个包含原始数据：
{2016: {1: 88698.0,
2: 86725.0,
3: 80426.0,
4: 74888.0,
5: 71659.0,
6: 67431.0,
7: 63613.0,
8: 60174.0,
9: 59495.0,
10: 59487.0,
11: 59118.0,
12: 59542.0,
13: 61170.0,
14: 63785.0,
15: 65038.0,
16: 67441.0,
17: 68188.0,
18: 69303.0,
19: 70224.0,
20: 70163.0,
21: 71522.0,
22: 73707.0,
23: 75002.0,
24: 76487.0,
25: 78806.0,
26: 81444.0,
27: 84114.0,
28: 84274.0,
29: 86701.0,
30: 87051.0,
31: 89298.0,
32: 91461.0,
33: 93937.0,
34: 96308.0,
35: 96803.0,
36: 98718.0,
37: 99343.0,
38: 100494.0,
39: 101260.0,
40: 101153.0,
41: 99668.0,
42: 97139.0,
43: 97203.0,
44: 95940.0,
45: 96969.0,
46: 98608.0,
47: 96332.0,
48: 94736.0,
49: 90970.0,
50: 87938.0,
51: 82082.0,
52: 79481.0,
53: nan},
2017: {1: 75212.0,
2: 68024.0,
3: 64087.0,
4: 58824.0,
5: 52226.0,
6: 50006.0,
7: 46975.0,
8: 46794.0,
9: 42855.0,
10: 42021.0,
11: 41884.0,
12: 40281.0,
13: 39117.0,
14: 37985.0,
15: 37120.0,
16: 36968.0,
17: 36702.0,
18: 38486.0,
19: 39051.0,
20: 40589.0,
21: 44099.0,
22: 47631.0,
23: 49984.0,
24: 51726.0,
25: 55653.0,
26: 57663.0,
27: 59409.0,
28: 62820.0,
29: 63324.0,
30: 64788.0,
31: 64693.0,
32: 66452.0,
33: 69349.0,
34: 70697.0,
35: 76470.0,
36: 78805.0,
37: 77624.0,
38: 75268.0,
39: 74695.0,
40: 75892.0,
41: 75930.0,
42: 74942.0,
43: 75824.0,
44: 74628.0,
45: 72058.0,
46: 71113.0,
47: 70602.0,
48: 71898.0,
49: 72186.0,
50: 68760.0,
51: 65931.0,
52: 65441.0,
53: nan},
2018: {1: 59224.0,
2: 55546.0,
3: 51355.0,
4: 50126.0,
5: 45962.0,
6: 42438.0,
7: 39840.0,
8: 39370.0,
9: 37844.0,
10: 35470.0,
11: 33731.0,
12: 32671.0,
13: 33416.0,
14: 33039.0,
15: 33260.0,
16: 32937.0,
17: 33599.0,
18: 35737.0,
19: 37453.0,
20: 38314.0,
21: 40159.0,
22: 44152.0,
23: 47971.0,
24: 51381.0,
25: 55825.0,
26: 58905.0,
27: 61242.0,
28: 62724.0,
29: 61766.0,
30: 63514.0,
31: 63533.0,
32: 66825.0,
33: 65732.0,
34: 68240.0,
35: 70572.0,
36: 71835.0,
37: 72966.0,
38: 74556.0,
39: 76592.0,
40: 78223.0,
41: 79895.0,
42: 79209.0,
43: 79793.0,
44: 80800.0,
45: 79795.0,
46: 78203.0,
47: 77027.0,
48: 75356.0,
49: 72124.0,
50: 68584.0,
51: 67402.0,
52: 65576.0,
53: nan},
2019: {1: 63624.0,
2: 62046.0,
3: 58091.0,
4: 54316.0,
5: 51765.0,
6: 52033.0,
7: 48140.0,
8: 46787.0,
9: 44772.0,
10: 43806.0,
11: 44905.0,
12: 45564.0,
13: 46906.0,
14: 48134.0,
15: 50554.0,
16: 51797.0,
17: 53271.0,
18: 54197.0,
19: 57114.0,
20: 60312.0,
21: 60509.0,
22: 63388.0,
23: 66265.0,
24: 69530.0,
25: 70905.0,
26: 72313.0,
27: 72288.0,
28: 73153.0,
29: 74967.0,
30: 76430.0,
31: 79261.0,
32: 82623.0,
33: 86492.0,
34: 90041.0,
35: 92856.0,
36: 93701.0,
37: 96520.0,
38: 95368.0,
39: 96264.0,
40: 96355.0,
41: 95794.0,
42: 95282.0,
43: 94817.0,
44: 95536.0,
45: 92914.0,
46: 89160.0,
47: 88321.0,
48: 86443.0,
49: 88099.0,
50: 85469.0,
51: 82634.0,
52: 82188.0,
53: nan},
2020: {1: 82784.0,
2: 81804.0,
3: 80581.0,
4: 77236.0,
5: 77976.0,
6: 71822.0,
7: 68726.0,
8: 68132.0,
9: 64557.0,
10: 61529.0,
11: 61379.0,
12: 59424.0,
13: 59134.0,
14: 59027.0,
15: 56780.0,
16: 57442.0,
17: 56835.0,
18: 59376.0,
19: 61625.0,
20: 62697.0,
21: 64240.0,
22: 67329.0,
23: 66282.0,
24: 68967.0,
25: 71331.0,
26: 74599.0,
27: 76823.0,
28: 80348.0,
29: 82388.0,
30: 84404.0,
31: 86713.0,
32: 89336.0,
33: 89295.0,
34: 90833.0,
35: 95222.0,
36: 97380.0,
37: 96141.0,
38: 97890.0,
39: 101959.0,
40: 101842.0,
41: 99897.0,
42: 98325.0,
43: 98391.0,
44: 95828.0,
45: 94889.0,
46: 92887.0,
47: 92562.0,
48: 91718.0,
49: 87637.0,
50: 83927.0,
51: 81596.0,
52: 75146.0,
53: 72777.0},
2021: {1: 66048.0,
2: 59818.0,
3: 57610.0,
4: 56053.0,
5: 51545.0,
6: 48649.0,
7: 43491.0,
8: 41246.0,
9: 41199.0,
10: 41029.0,
11: 41269.0,
12: nan,
13: nan,
14: nan,
15: nan,
16: nan,
17: nan,
18: nan,
19: nan,
20: nan,
21: nan,
22: nan,
23: nan,
24: nan,
25: nan,
26: nan,
27: nan,
28: nan,
29: nan,
30: nan,
31: nan,
32: nan,
33: nan,
34: nan,
35: nan,
36: nan,
37: nan,
38: nan,
39: nan,
40: nan,
41: nan,
42: nan,
43: nan,
44: nan,
45: nan,
46: nan,
47: nan,
48: nan,
49: nan,
50: nan,
51: nan,
52: nan,
53: nan}}
然后是第一个数据帧。diff((：
{2016: {1: nan,
2: -1973.0,
3: -6299.0,
4: -5538.0,
5: -3229.0,
6: -4228.0,
7: -3818.0,
8: -3439.0,
9: -679.0,
10: -8.0,
11: -369.0,
12: 424.0,
13: 1628.0,
14: 2615.0,
15: 1253.0,
16: 2403.0,
17: 747.0,
18: 1115.0,
19: 921.0,
20: -61.0,
21: 1359.0,
22: 2185.0,
23: 1295.0,
24: 1485.0,
25: 2319.0,
26: 2638.0,
27: 2670.0,
28: 160.0,
29: 2427.0,
30: 350.0,
31: 2247.0,
32: 2163.0,
33: 2476.0,
34: 2371.0,
35: 495.0,
36: 1915.0,
37: 625.0,
38: 1151.0,
39: 766.0,
40: -107.0,
41: -1485.0,
42: -2529.0,
43: 64.0,
44: -1263.0,
45: 1029.0,
46: 1639.0,
47: -2276.0,
48: -1596.0,
49: -3766.0,
50: -3032.0,
51: -5856.0,
52: -2601.0,
53: nan},
2017: {1: nan,
2: -7188.0,
3: -3937.0,
4: -5263.0,
5: -6598.0,
6: -2220.0,
7: -3031.0,
8: -181.0,
9: -3939.0,
10: -834.0,
11: -137.0,
12: -1603.0,
13: -1164.0,
14: -1132.0,
15: -865.0,
16: -152.0,
17: -266.0,
18: 1784.0,
19: 565.0,
20: 1538.0,
21: 3510.0,
22: 3532.0,
23: 2353.0,
24: 1742.0,
25: 3927.0,
26: 2010.0,
27: 1746.0,
28: 3411.0,
29: 504.0,
30: 1464.0,
31: -95.0,
32: 1759.0,
33: 2897.0,
34: 1348.0,
35: 5773.0,
36: 2335.0,
37: -1181.0,
38: -2356.0,
39: -573.0,
40: 1197.0,
41: 38.0,
42: -988.0,
43: 882.0,
44: -1196.0,
45: -2570.0,
46: -945.0,
47: -511.0,
48: 1296.0,
49: 288.0,
50: -3426.0,
51: -2829.0,
52: -490.0,
53: nan},
2018: {1: nan,
2: -3678.0,
3: -4191.0,
4: -1229.0,
5: -4164.0,
6: -3524.0,
7: -2598.0,
8: -470.0,
9: -1526.0,
10: -2374.0,
11: -1739.0,
12: -1060.0,
13: 745.0,
14: -377.0,
15: 221.0,
16: -323.0,
17: 662.0,
18: 2138.0,
19: 1716.0,
20: 861.0,
21: 1845.0,
22: 3993.0,
23: 3819.0,
24: 3410.0,
25: 4444.0,
26: 3080.0,
27: 2337.0,
28: 1482.0,
29: -958.0,
30: 1748.0,
31: 19.0,
32: 3292.0,
33: -1093.0,
34: 2508.0,
35: 2332.0,
36: 1263.0,
37: 1131.0,
38: 1590.0,
39: 2036.0,
40: 1631.0,
41: 1672.0,
42: -686.0,
43: 584.0,
44: 1007.0,
45: -1005.0,
46: -1592.0,
47: -1176.0,
48: -1671.0,
49: -3232.0,
50: -3540.0,
51: -1182.0,
52: -1826.0,
53: nan},
2019: {1: nan,
2: -1578.0,
3: -3955.0,
4: -3775.0,
5: -2551.0,
6: 268.0,
7: -3893.0,
8: -1353.0,
9: -2015.0,
10: -966.0,
11: 1099.0,
12: 659.0,
13: 1342.0,
14: 1228.0,
15: 2420.0,
16: 1243.0,
17: 1474.0,
18: 926.0,
19: 2917.0,
20: 3198.0,
21: 197.0,
22: 2879.0,
23: 2877.0,
24: 3265.0,
25: 1375.0,
26: 1408.0,
27: -25.0,
28: 865.0,
29: 1814.0,
30: 1463.0,
31: 2831.0,
32: 3362.0,
33: 3869.0,
34: 3549.0,
35: 2815.0,
36: 845.0,
37: 2819.0,
38: -1152.0,
39: 896.0,
40: 91.0,
41: -561.0,
42: -512.0,
43: -465.0,
44: 719.0,
45: -2622.0,
46: -3754.0,
47: -839.0,
48: -1878.0,
49: 1656.0,
50: -2630.0,
51: -2835.0,
52: -446.0,
53: nan},
2020: {1: nan,
2: -980.0,
3: -1223.0,
4: -3345.0,
5: 740.0,
6: -6154.0,
7: -3096.0,
8: -594.0,
9: -3575.0,
10: -3028.0,
11: -150.0,
12: -1955.0,
13: -290.0,
14: -107.0,
15: -2247.0,
16: 662.0,
17: -607.0,
18: 2541.0,
19: 2249.0,
20: 1072.0,
21: 1543.0,
22: 3089.0,
23: -1047.0,
24: 2685.0,
25: 2364.0,
26: 3268.0,
27: 2224.0,
28: 3525.0,
29: 2040.0,
30: 2016.0,
31: 2309.0,
32: 2623.0,
33: -41.0,
34: 1538.0,
35: 4389.0,
36: 2158.0,
37: -1239.0,
38: 1749.0,
39: 4069.0,
40: -117.0,
41: -1945.0,
42: -1572.0,
43: 66.0,
44: -2563.0,
45: -939.0,
46: -2002.0,
47: -325.0,
48: -844.0,
49: -4081.0,
50: -3710.0,
51: -2331.0,
52: -6450.0,
53: -2369.0}}
我试图做的是，对于2021为NaN的任何行中的所有列，通过取正常数据帧中的值并添加.diff((数据帧中下一个向下的值来计算下一个值行。因此，例如，第12周的2020将是61379(正常df中的第11行(+(-195.0.diff(df(中的第12行(
TIA-
与之前的逻辑相同
out = df1.mask(df1[2021].notna(),df1+df2.shift(-1),axis=0).fillna(df1[[2021]])