Python生命线- ConvergenceWarning:牛顿-拉夫森未能在考克斯道具风险中充分收敛。
Python生命线- ConvergenceWarning:牛顿-拉夫森未能在考克斯道具风险中充分收敛。
提问于 2019-07-30 08:25:52
在我的完整数据集上调用CoxPHFitter()时，我得到以下错误：
Users/anaconda3/lib/python3.7/site-packages/lifelines/fitters/coxph_fitter.py:557: ConvergenceWarning: Newton-Rhapson failed to converge sufficiently in 50 steps.
warnings.warn("Newton-Rhapson failed to converge sufficiently in %d steps." % max_steps, ConvergenceWarning)但我找不到增加步数的方法。我还试着使用params的价值观：step_size、enalizer和alpha --但没有成功。
这是我正在运行的函数和params：
def cox_proportional_hazard_model(data, survival_duration, survival_status, strata=None):
cph = CoxPHFitter(alpha=0.05, tie_method='Efron', penalizer=0.1, strata=None)
cph.fit(df=data,
duration_col=survival_duration, event_col=survival_status,
strata=strata, show_progress=True, step_size=0.1)
cph.print_summary()
return cph这是输出和三角洲：
Iteration 1: norm_delta = 22.95175, step_size = 0.1000, ll = -383.78983, newton_decrement = 224.62787, seconds_since_start = 0.0
Iteration 2: norm_delta = 8.59969, step_size = 0.0250, ll = -344.73687, newton_decrement = 100.90631, seconds_since_start = 0.1
Iteration 3: norm_delta = 8.00526, step_size = 0.0225, ll = -339.71541, newton_decrement = 95.05309, seconds_since_start = 0.1
Iteration 4: norm_delta = 7.61510, step_size = 0.0243, ll = -335.44970, newton_decrement = 90.63796, seconds_since_start = 0.1
Iteration 5: norm_delta = 7.29316, step_size = 0.0262, ll = -331.05741, newton_decrement = 86.53240, seconds_since_start = 0.2
Iteration 6: norm_delta = 7.02757, step_size = 0.0283, ll = -326.52929, newton_decrement = 82.69935, seconds_since_start = 0.2
Iteration 7: norm_delta = 6.80949, step_size = 0.0306, ll = -321.85618, newton_decrement = 79.10759, seconds_since_start = 0.2
Iteration 8: norm_delta = 6.63229, step_size = 0.0331, ll = -317.02892, newton_decrement = 75.73047, seconds_since_start = 0.3
Iteration 9: norm_delta = 6.49106, step_size = 0.0357, ll = -312.03837, newton_decrement = 72.54478, seconds_since_start = 0.3
Iteration 10: norm_delta = 6.38213, step_size = 0.0386, ll = -306.87533, newton_decrement = 69.52988, seconds_since_start = 0.3
Iteration 11: norm_delta = 6.30276, step_size = 0.0416, ll = -301.53059, newton_decrement = 66.66703, seconds_since_start = 0.3
Iteration 12: norm_delta = 6.25096, step_size = 0.0450, ll = -295.99496, newton_decrement = 63.93876, seconds_since_start = 0.4
Iteration 13: norm_delta = 6.22523, step_size = 0.0486, ll = -290.25932, newton_decrement = 61.32840, seconds_since_start = 0.4
Iteration 14: norm_delta = 6.22451, step_size = 0.0525, ll = -284.31480, newton_decrement = 58.81965, seconds_since_start = 0.4
Iteration 15: norm_delta = 6.24798, step_size = 0.0567, ll = -278.15291, newton_decrement = 56.39613, seconds_since_start = 0.5
Iteration 16: norm_delta = 6.29497, step_size = 0.0500, ll = -271.76578, newton_decrement = 54.04106, seconds_since_start = 0.5
Iteration 17: norm_delta = 6.35101, step_size = 0.0441, ll = -266.33229, newton_decrement = 52.14109, seconds_since_start = 0.5
Iteration 18: norm_delta = 6.40935, step_size = 0.0389, ll = -261.68076, newton_decrement = 50.57648, seconds_since_start = 0.6
Iteration 19: norm_delta = 6.46639, step_size = 0.0343, ll = -257.67951, newton_decrement = 49.26847, seconds_since_start = 0.6
Iteration 20: norm_delta = 6.52023, step_size = 0.0302, ll = -254.22460, newton_decrement = 48.16262, seconds_since_start = 0.6
Iteration 21: norm_delta = 6.57000, step_size = 0.0267, ll = -251.23237, newton_decrement = 47.21966, seconds_since_start = 0.7
Iteration 22: norm_delta = 6.61537, step_size = 0.0235, ll = -248.63436, newton_decrement = 46.41034, seconds_since_start = 0.7
Iteration 23: norm_delta = 6.65633, step_size = 0.0207, ll = -246.37393, newton_decrement = 45.71217, seconds_since_start = 0.7
Iteration 24: norm_delta = 6.69308, step_size = 0.0183, ll = -244.40374, newton_decrement = 45.10750, seconds_since_start = 0.8
Iteration 25: norm_delta = 6.72589, step_size = 0.0161, ll = -242.68394, newton_decrement = 44.58214, seconds_since_start = 0.8
Iteration 26: norm_delta = 6.75508, step_size = 0.0142, ll = -241.18077, newton_decrement = 44.12455, seconds_since_start = 0.8
Iteration 27: norm_delta = 6.78100, step_size = 0.0126, ll = -239.86547, newton_decrement = 43.72517, seconds_since_start = 0.9
Iteration 28: norm_delta = 6.80396, step_size = 0.0111, ll = -238.71344, newton_decrement = 43.37602, seconds_since_start = 0.9
Iteration 29: norm_delta = 6.82427, step_size = 0.0098, ll = -237.70355, newton_decrement = 43.07037, seconds_since_start = 0.9
Iteration 30: norm_delta = 6.84223, step_size = 0.0086, ll = -236.81763, newton_decrement = 42.80250, seconds_since_start = 1.0
Iteration 31: norm_delta = 6.85809, step_size = 0.0076, ll = -236.03994, newton_decrement = 42.56752, seconds_since_start = 1.0
Iteration 32: norm_delta = 6.87209, step_size = 0.0067, ll = -235.35688, newton_decrement = 42.36124, seconds_since_start = 1.0
Iteration 33: norm_delta = 6.88445, step_size = 0.0059, ll = -234.75663, newton_decrement = 42.18003, seconds_since_start = 1.1
Iteration 34: norm_delta = 6.89535, step_size = 0.0052, ll = -234.22893, newton_decrement = 42.02075, seconds_since_start = 1.1
Iteration 35: norm_delta = 6.90497, step_size = 0.0046, ll = -233.76482, newton_decrement = 41.88069, seconds_since_start = 1.1
Iteration 36: norm_delta = 6.91345, step_size = 0.0041, ll = -233.35650, newton_decrement = 41.75748, seconds_since_start = 1.2
Iteration 37: norm_delta = 6.92093, step_size = 0.0036, ll = -232.99717, newton_decrement = 41.64906, seconds_since_start = 1.2
Iteration 38: norm_delta = 6.92753, step_size = 0.0032, ll = -232.68085, newton_decrement = 41.55361, seconds_since_start = 1.2
Iteration 39: norm_delta = 6.93335, step_size = 0.0028, ll = -232.40235, newton_decrement = 41.46957, seconds_since_start = 1.3
Iteration 40: norm_delta = 6.93847, step_size = 0.0025, ll = -232.15707, newton_decrement = 41.39556, seconds_since_start = 1.3
Iteration 41: norm_delta = 6.94300, step_size = 0.0022, ll = -231.94104, newton_decrement = 41.33037, seconds_since_start = 1.3
Iteration 42: norm_delta = 6.94698, step_size = 0.0019, ll = -231.75071, newton_decrement = 41.27294, seconds_since_start = 1.4
Iteration 43: norm_delta = 6.95050, step_size = 0.0017, ll = -231.58303, newton_decrement = 41.22233, seconds_since_start = 1.4
Iteration 44: norm_delta = 6.95360, step_size = 0.0015, ll = -231.43526, newton_decrement = 41.17774, seconds_since_start = 1.4
Iteration 45: norm_delta = 6.95634, step_size = 0.0013, ll = -231.30504, newton_decrement = 41.13844, seconds_since_start = 1.5
Iteration 46: norm_delta = 6.95875, step_size = 0.0012, ll = -231.19026, newton_decrement = 41.10379, seconds_since_start = 1.5
Iteration 47: norm_delta = 6.96087, step_size = 0.0010, ll = -231.08909, newton_decrement = 41.07326, seconds_since_start = 1.5
Iteration 48: norm_delta = 6.96275, step_size = 0.0009, ll = -230.99991, newton_decrement = 41.04634, seconds_since_start = 1.6
Iteration 49: norm_delta = 6.96440, step_size = 0.0008, ll = -230.92129, newton_decrement = 41.02261, seconds_since_start = 1.6
Iteration 50: norm_delta = 6.96586, step_size = 0.0007, ll = -230.85198, newton_decrement = 41.00169, seconds_since_start = 1.7
Convergence failed. See any warning messages.
Concordance index of the model 0.9980554205153136
<lifelines.CoxPHFitter: fitted with 115 observations, 19 censored>
duration col = 'Survival from onset'
event col = 'survival status'
penalizer = 0.1
number of subjects = 115
number of events = 96
log-likelihood = -230.85
time fit was run = 2019-07-29 18:06:24 UTC
---
coef exp(coef) se(coef) z p -log2(p) lower 0.95 upper 0.95
hsa-miR-1-3p 0.00 1.00 0.00 0.10 0.92 0.12 -0.00 0.00
hsa-miR-101-3p 0.00 1.00 0.00 0.35 0.73 0.45 -0.00 0.01
hsa-miR-103a-3p 0.00 1.00 0.00 0.78 0.44 1.19 -0.00 0.00
hsa-miR-103b -0.00 1.00 0.00 -0.35 0.73 0.46 -0.01 0.01
hsa-miR-106b-3p -0.00 1.00 0.01 -0.10 0.92 0.12 -0.02 0.01
hsa-miR-107 0.00 1.00 0.01 0.11 0.91 0.13 -0.01 0.01
hsa-miR-10a-5p -0.00 1.00 0.02 -0.20 0.84 0.25 -0.04 0.03
hsa-miR-10b-5p -0.01 0.99 0.02 -0.34 0.73 0.45 -0.04 0.02
hsa-miR-122-5p 0.00 1.00 0.00 0.05 0.96 0.06 -0.00 0.00
hsa-miR-125a-5p -0.00 1.00 0.00 -0.22 0.83 0.27 -0.01 0.01
hsa-miR-125b-2-3p -0.00 1.00 0.01 -0.17 0.86 0.21 -0.01 0.01
hsa-miR-125b-5p -0.00 1.00 0.01 -0.29 0.77 0.38 -0.01 0.01
hsa-miR-126-3p 0.00 1.00 0.00 0.09 0.93 0.11 -0.00 0.00
hsa-miR-126-5p 0.00 1.00 0.00 0.29 0.77 0.38 -0.00 0.00
hsa-miR-1268b 0.01 1.01 0.02 0.41 0.68 0.56 -0.04 0.06
hsa-miR-127-3p 0.01 1.01 0.03 0.46 0.64 0.64 -0.04 0.07
hsa-miR-128-3p -0.00 1.00 0.01 -0.12 0.91 0.14 -0.03 0.03
hsa-miR-1287-5p -0.00 1.00 0.02 -0.02 0.98 0.03 -0.04 0.04
hsa-miR-1301-3p 0.00 1.00 0.03 0.03 0.98 0.03 -0.05 0.05
hsa-miR-1306-5p -0.01 0.99 0.02 -0.33 0.74 0.44 -0.05 0.04
hsa-miR-1307-3p -0.00 1.00 0.00 -0.21 0.83 0.27 -0.01 0.01
hsa-miR-1307-5p 0.00 1.00 0.03 0.04 0.96 0.05 -0.06 0.06
hsa-miR-130a-3p 0.01 1.01 0.03 0.22 0.82 0.28 -0.05 0.06
hsa-miR-133a-3p 0.00 1.00 0.00 0.08 0.94 0.09 -0.01 0.01
hsa-miR-133b -0.00 1.00 0.02 -0.17 0.87 0.21 -0.03 0.03
hsa-miR-134-5p -0.00 1.00 0.01 -0.26 0.80 0.33 -0.01 0.01
hsa-miR-139-3p 0.00 1.00 0.02 0.06 0.95 0.08 -0.03 0.03
hsa-miR-140-3p 0.00 1.00 0.00 0.20 0.84 0.25 -0.01 0.01
hsa-miR-140-5p 0.02 1.02 0.05 0.33 0.74 0.43 -0.09 0.12
hsa-miR-142-3p -0.00 1.00 0.00 -0.03 0.97 0.04 -0.00 0.00
hsa-miR-142-5p 0.00 1.00 0.00 0.01 0.99 0.01 -0.00 0.00
hsa-miR-143-3p -0.00 1.00 0.00 -0.09 0.93 0.10 -0.00 0.00
hsa-miR-144-3p -0.00 1.00 0.01 -0.21 0.83 0.26 -0.03 0.02
hsa-miR-144-5p -0.00 1.00 0.03 -0.09 0.93 0.11 -0.07 0.06
hsa-miR-145-5p 0.01 1.01 0.04 0.26 0.80 0.33 -0.07 0.09
hsa-miR-146a-5p 0.00 1.00 0.00 0.01 0.99 0.01 -0.00 0.00
hsa-miR-146b-5p 0.00 1.00 0.01 0.04 0.97 0.05 -0.02 0.02
hsa-miR-148a-3p 0.00 1.00 0.00 0.54 0.59 0.76 -0.00 0.00
hsa-miR-148b-3p -0.00 1.00 0.00 -0.50 0.62 0.69 -0.01 0.01
hsa-miR-150-5p -0.00 1.00 0.00 -0.25 0.81 0.31 -0.00 0.00
hsa-miR-151a-3p -0.00 1.00 0.00 -0.09 0.93 0.10 -0.00 0.00
hsa-miR-151b/151a-5p 0.00 1.00 0.02 0.22 0.82 0.28 -0.03 0.04
hsa-miR-152-3p -0.00 1.00 0.02 -0.20 0.84 0.25 -0.04 0.03
hsa-miR-155-5p 0.00 1.00 0.01 0.58 0.56 0.83 -0.01 0.01
hsa-miR-15a-5p 0.00 1.00 0.02 0.15 0.88 0.19 -0.03 0.03
hsa-miR-15b-5p -0.00 1.00 0.01 -0.11 0.91 0.13 -0.01 0.01
hsa-miR-16-2-3p -0.00 1.00 0.01 -0.26 0.80 0.33 -0.02 0.01
hsa-miR-16-5p 0.00 1.00 0.00 0.15 0.88 0.18 -0.00 0.00
hsa-miR-17-5p -0.01 0.99 0.02 -0.32 0.75 0.41 -0.04 0.03
hsa-miR-181a-2-3p -0.01 0.99 0.07 -0.08 0.94 0.09 -0.14 0.13
hsa-miR-181a-5p 0.00 1.00 0.00 0.13 0.90 0.16 -0.01 0.01
hsa-miR-181b-5p 0.02 1.02 0.03 0.67 0.51 0.98 -0.04 0.07
hsa-miR-182-5p -0.00 1.00 0.01 -0.02 0.98 0.03 -0.02 0.02
hsa-miR-183-5p 0.00 1.00 0.01 0.22 0.83 0.27 -0.02 0.02
hsa-miR-185-3p 0.00 1.00 0.05 0.09 0.93 0.10 -0.09 0.10
hsa-miR-185-5p 0.00 1.00 0.00 0.14 0.89 0.17 -0.01 0.01
hsa-miR-186-5p -0.00 1.00 0.01 -0.03 0.97 0.04 -0.02 0.02
hsa-miR-18a-5p 0.02 1.02 0.06 0.32 0.75 0.42 -0.09 0.13
hsa-miR-1908-5p -0.00 1.00 0.03 -0.14 0.89 0.16 -0.06 0.05
hsa-miR-190a-5p 0.00 1.00 0.04 0.13 0.90 0.15 -0.07 0.08
hsa-miR-191-5p -0.00 1.00 0.00 -0.47 0.64 0.65 -0.00 0.00
hsa-miR-192-5p 0.00 1.00 0.01 0.10 0.92 0.12 -0.02 0.02
hsa-miR-193a-5p -0.00 1.00 0.03 -0.11 0.91 0.14 -0.06 0.05
hsa-miR-194-5p -0.00 1.00 0.04 -0.09 0.93 0.11 -0.08 0.07
hsa-miR-195-5p -0.00 1.00 0.01 -0.28 0.78 0.36 -0.02 0.02
hsa-miR-196b-5p -0.01 0.99 0.02 -0.63 0.53 0.92 -0.06 0.03
hsa-miR-197-3p -0.00 1.00 0.02 -0.14 0.89 0.18 -0.04 0.03
hsa-miR-199a-3p -0.00 1.00 0.00 -0.54 0.59 0.76 -0.00 0.00
hsa-miR-199a-5p 0.02 1.02 0.04 0.42 0.68 0.56 -0.06 0.09
hsa-miR-199b-3p -0.00 1.00 0.00 -0.41 0.68 0.55 -0.01 0.01
hsa-miR-19a-3p 0.00 1.00 0.04 0.03 0.97 0.04 -0.08 0.09
hsa-miR-19b-3p 0.01 1.01 0.02 0.35 0.73 0.46 -0.03 0.04
hsa-miR-200a-3p 0.00 1.00 0.04 0.04 0.97 0.05 -0.07 0.07
hsa-miR-200b-3p 0.01 1.01 0.02 0.48 0.63 0.66 -0.03 0.04
hsa-miR-200c-3p 0.00 1.00 0.04 0.02 0.98 0.02 -0.08 0.08
hsa-miR-203a-3p -0.00 1.00 0.02 -0.24 0.81 0.30 -0.03 0.03
hsa-miR-205-5p -0.01 0.99 0.03 -0.18 0.86 0.22 -0.06 0.05
hsa-miR-206 0.00 1.00 0.00 0.13 0.90 0.16 -0.00 0.00
hsa-miR-20a-5p -0.00 1.00 0.01 -0.14 0.89 0.17 -0.02 0.02
hsa-miR-20b-5p -0.02 0.98 0.04 -0.45 0.65 0.61 -0.09 0.05
hsa-miR-21-5p 0.00 1.00 0.00 0.02 0.98 0.03 -0.00 0.00
hsa-miR-2110 -0.02 0.98 0.05 -0.39 0.70 0.52 -0.12 0.08
hsa-miR-22-3p 0.00 1.00 0.01 0.49 0.62 0.69 -0.01 0.02
hsa-miR-221-3p -0.00 1.00 0.00 -0.07 0.95 0.08 -0.00 0.00
hsa-miR-222-3p -0.00 1.00 0.01 -0.13 0.90 0.16 -0.02 0.02
hsa-miR-223-3p 0.00 1.00 0.00 0.46 0.64 0.64 -0.00 0.00
hsa-miR-223-5p 0.00 1.00 0.01 0.12 0.91 0.14 -0.03 0.03
hsa-miR-23a-3p 0.00 1.00 0.00 0.03 0.98 0.03 -0.01 0.01
hsa-miR-23b-3p 0.01 1.01 0.01 0.40 0.69 0.53 -0.02 0.03
hsa-miR-24-3p -0.00 1.00 0.00 -0.03 0.97 0.04 -0.01 0.01
hsa-miR-25-3p -0.00 1.00 0.00 -0.01 1.00 0.01 -0.00 0.00
hsa-miR-26a-5p 0.00 1.00 0.00 0.05 0.96 0.06 -0.00 0.00
hsa-miR-26b-5p -0.00 1.00 0.00 -0.17 0.87 0.20 -0.00 0.00
hsa-miR-27a-3p -0.00 1.00 0.01 -0.27 0.78 0.35 -0.03 0.02
hsa-miR-27b-3p -0.00 1.00 0.01 -0.21 0.83 0.26 -0.01 0.01
hsa-miR-28-3p 0.00 1.00 0.01 0.05 0.96 0.06 -0.02 0.02
hsa-miR-29a-3p -0.00 1.00 0.00 -0.58 0.56 0.82 -0.01 0.00
hsa-miR-29b-3p -0.00 1.00 0.03 -0.15 0.88 0.19 -0.06 0.05
hsa-miR-29c-3p -0.00 1.00 0.01 -0.18 0.86 0.22 -0.01 0.01
hsa-miR-29c-5p -0.00 1.00 0.05 -0.02 0.98 0.02 -0.09 0.09
hsa-miR-30a-5p -0.00 1.00 0.01 -0.10 0.92 0.12 -0.01 0.01
hsa-miR-30b-5p -0.01 0.99 0.04 -0.12 0.90 0.15 -0.09 0.08
hsa-miR-30c-5p 0.00 1.00 0.02 0.15 0.88 0.19 -0.03 0.04
hsa-miR-30d-5p 0.00 1.00 0.00 0.19 0.85 0.24 -0.00 0.00
hsa-miR-30e-3p -0.00 1.00 0.01 -0.14 0.89 0.18 -0.02 0.02
hsa-miR-30e-5p -0.00 1.00 0.00 -0.55 0.59 0.77 -0.01 0.00
hsa-miR-3135b 0.00 1.00 0.01 0.14 0.89 0.17 -0.02 0.03
hsa-miR-3168 0.00 1.00 0.00 0.30 0.76 0.39 -0.00 0.00
hsa-miR-320a -0.00 1.00 0.00 -0.04 0.97 0.04 -0.00 0.00
hsa-miR-320b -0.00 1.00 0.01 -0.23 0.82 0.29 -0.03 0.02
hsa-miR-320c -0.00 1.00 0.02 -0.14 0.89 0.17 -0.05 0.04
hsa-miR-323b-3p 0.04 1.04 0.05 0.72 0.47 1.09 -0.06 0.13
hsa-miR-324-5p -0.01 0.99 0.04 -0.16 0.88 0.19 -0.07 0.06
hsa-miR-326 -0.00 1.00 0.03 -0.12 0.90 0.15 -0.06 0.05
hsa-miR-328-3p -0.00 1.00 0.01 -0.06 0.95 0.07 -0.01 0.01
hsa-miR-335-5p -0.01 0.99 0.02 -0.50 0.62 0.69 -0.04 0.03
hsa-miR-339-3p -0.00 1.00 0.04 -0.05 0.96 0.05 -0.07 0.07
hsa-miR-339-5p 0.00 1.00 0.01 0.42 0.67 0.57 -0.01 0.02
hsa-miR-340-5p 0.01 1.01 0.02 0.34 0.73 0.45 -0.04 0.05
hsa-miR-342-3p -0.00 1.00 0.00 -0.05 0.96 0.06 -0.00 0.00
hsa-miR-345-5p 0.00 1.00 0.05 0.02 0.98 0.03 -0.09 0.10
hsa-miR-34a-5p 0.00 1.00 0.03 0.16 0.87 0.20 -0.05 0.06
hsa-miR-361-3p 0.01 1.01 0.03 0.44 0.66 0.61 -0.04 0.06
hsa-miR-361-5p 0.00 1.00 0.01 0.38 0.70 0.50 -0.02 0.03
hsa-miR-3613-3p 0.01 1.01 0.02 0.32 0.75 0.41 -0.03 0.04
hsa-miR-3615 0.00 1.00 0.01 0.24 0.81 0.30 -0.02 0.03
hsa-miR-363-3p 0.01 1.01 0.02 0.44 0.66 0.60 -0.03 0.05
hsa-miR-3687 0.02 1.02 0.04 0.62 0.54 0.90 -0.05 0.10
hsa-miR-370-3p -0.00 1.00 0.02 -0.11 0.91 0.14 -0.03 0.03
hsa-miR-374a-5p -0.01 0.99 0.04 -0.37 0.71 0.49 -0.09 0.06
hsa-miR-374b-5p -0.02 0.98 0.08 -0.29 0.77 0.38 -0.17 0.13
hsa-miR-375 0.00 1.00 0.01 0.07 0.94 0.09 -0.02 0.02
hsa-miR-378a-3p 0.00 1.00 0.01 0.11 0.92 0.13 -0.02 0.03
hsa-miR-378c 0.02 1.02 0.04 0.40 0.69 0.53 -0.06 0.10
hsa-miR-379-5p 0.00 1.00 0.02 0.21 0.83 0.26 -0.03 0.04
hsa-miR-381-3p 0.00 1.00 0.04 0.12 0.90 0.15 -0.07 0.08
hsa-miR-382-5p 0.00 1.00 0.01 0.19 0.85 0.24 -0.01 0.01
hsa-miR-3940-3p -0.01 0.99 0.02 -0.26 0.79 0.33 -0.04 0.03
hsa-miR-3974 -0.00 1.00 0.02 -0.21 0.83 0.26 -0.03 0.03
hsa-miR-409-3p -0.00 1.00 0.01 -0.48 0.63 0.67 -0.01 0.01
hsa-miR-423-3p -0.00 1.00 0.01 -0.72 0.47 1.08 -0.01 0.01
hsa-miR-423-5p 0.00 1.00 0.00 0.15 0.88 0.19 -0.00 0.00
hsa-miR-425-3p 0.00 1.00 0.02 0.08 0.94 0.09 -0.04 0.05
hsa-miR-425-5p -0.00 1.00 0.00 -0.21 0.84 0.26 -0.00 0.00
hsa-miR-4254 -0.00 1.00 0.01 -0.30 0.77 0.38 -0.03 0.02
hsa-miR-4286 0.03 1.03 0.06 0.56 0.57 0.80 -0.08 0.14
hsa-miR-431-5p -0.00 1.00 0.02 -0.15 0.88 0.18 -0.04 0.03
hsa-miR-432-5p 0.00 1.00 0.00 0.28 0.78 0.35 -0.00 0.01
hsa-miR-4433a-3p -0.00 1.00 0.03 -0.01 0.99 0.01 -0.06 0.06
hsa-miR-4433b-5p 0.00 1.00 0.01 0.20 0.84 0.25 -0.01 0.01
hsa-miR-4446-3p 0.01 1.01 0.04 0.24 0.81 0.30 -0.07 0.09
hsa-miR-4451 -0.00 1.00 0.02 -0.00 1.00 0.00 -0.04 0.04
hsa-miR-4454 -0.01 0.99 0.03 -0.28 0.78 0.36 -0.08 0.06
hsa-miR-451a 0.00 1.00 0.00 0.23 0.82 0.29 -0.00 0.00
hsa-miR-454-3p -0.00 1.00 0.02 -0.12 0.91 0.14 -0.05 0.04
hsa-miR-4655-5p -0.00 1.00 0.01 -0.05 0.96 0.06 -0.02 0.02
hsa-miR-4732-5p 0.00 1.00 0.02 0.02 0.98 0.03 -0.03 0.03
hsa-miR-483-5p 0.01 1.01 0.01 0.50 0.62 0.70 -0.02 0.03
hsa-miR-484 0.00 1.00 0.00 0.11 0.91 0.13 -0.01 0.01
hsa-miR-485-3p -0.00 1.00 0.02 -0.11 0.91 0.14 -0.03 0.03
hsa-miR-485-5p 0.01 1.01 0.05 0.16 0.87 0.19 -0.10 0.12
hsa-miR-486-3p 0.00 1.00 0.00 0.13 0.89 0.16 -0.00 0.00
hsa-miR-486-5p 0.00 1.00 0.00 0.09 0.93 0.11 -0.00 0.00
hsa-miR-487b-3p -0.00 1.00 0.05 -0.04 0.97 0.05 -0.10 0.09
hsa-miR-501-3p 0.02 1.02 0.04 0.42 0.68 0.57 -0.07 0.10
hsa-miR-532-5p -0.01 0.99 0.04 -0.34 0.73 0.45 -0.08 0.06
hsa-miR-548ad-3p 0.00 1.00 0.02 0.17 0.87 0.21 -0.03 0.04
hsa-miR-548ap-5p/548j-5p -0.00 1.00 0.04 -0.10 0.92 0.12 -0.09 0.08
hsa-miR-574-3p 0.01 1.01 0.02 0.25 0.80 0.32 -0.03 0.04
hsa-miR-584-5p 0.00 1.00 0.00 0.53 0.60 0.74 -0.00 0.01
hsa-miR-625-3p -0.00 1.00 0.00 -0.03 0.97 0.04 -0.01 0.01
hsa-miR-625-5p 0.01 1.01 0.02 0.29 0.77 0.37 -0.03 0.04
hsa-miR-628-3p -0.00 1.00 0.03 -0.03 0.97 0.04 -0.07 0.07
hsa-miR-629-5p -0.00 1.00 0.02 -0.21 0.83 0.26 -0.03 0.03
hsa-miR-652-3p 0.00 1.00 0.04 0.11 0.92 0.13 -0.08 0.09
hsa-miR-654-3p -0.01 0.99 0.02 -0.26 0.80 0.33 -0.05 0.04
hsa-miR-660-5p -0.01 0.99 0.02 -0.76 0.45 1.16 -0.05 0.02
hsa-miR-664a-5p -0.01 0.99 0.03 -0.26 0.79 0.34 -0.06 0.05
hsa-miR-671-3p -0.01 0.99 0.06 -0.17 0.87 0.21 -0.13 0.11
hsa-miR-671-5p 0.02 1.02 0.05 0.29 0.77 0.38 -0.09 0.12
hsa-miR-6728-5p -0.00 1.00 0.01 -0.13 0.89 0.16 -0.03 0.02
hsa-miR-6749-5p -0.01 0.99 0.01 -0.50 0.62 0.70 -0.04 0.02
hsa-miR-6787-5p -0.00 1.00 0.03 -0.19 0.85 0.24 -0.06 0.05
hsa-miR-6852-5p -0.01 0.99 0.06 -0.15 0.88 0.19 -0.12 0.10
hsa-miR-6890-5p 0.00 1.00 0.02 0.10 0.92 0.12 -0.04 0.05
hsa-miR-7-5p 0.01 1.01 0.02 0.38 0.70 0.51 -0.03 0.05
hsa-miR-744-5p -0.00 1.00 0.00 -0.07 0.95 0.08 -0.01 0.01
hsa-miR-760 0.03 1.03 0.05 0.49 0.62 0.69 -0.08 0.13
hsa-miR-769-5p 0.00 1.00 0.06 0.02 0.99 0.02 -0.11 0.11
hsa-miR-92a-3p 0.00 1.00 0.00 0.17 0.87 0.20 -0.00 0.00
hsa-miR-92b-3p -0.00 1.00 0.01 -0.08 0.94 0.09 -0.02 0.02
hsa-miR-93-5p -0.00 1.00 0.00 -0.30 0.76 0.39 -0.00 0.00
hsa-miR-941 -0.03 0.97 0.04 -0.65 0.51 0.96 -0.12 0.06
hsa-miR-98-5p 0.01 1.01 0.01 0.51 0.61 0.71 -0.02 0.03
hsa-miR-99a-5p 0.00 1.00 0.02 0.23 0.82 0.28 -0.04 0.05
hsa-miR-99b-5p 0.00 1.00 0.01 0.13 0.90 0.15 -0.01 0.01
---
Concordance = 1.00
Log-likelihood ratio test = 305.88 on 196 df, -log2(p)=20.20
/Users/nancy/anaconda3/lib/python3.7/site-packages/lifelines/fitters/coxph_fitter.py:557: ConvergenceWarning: Newton-Rhapson failed to converge sufficiently in 50 steps.
warnings.warn("Newton-Rhapson failed to converge sufficiently in %d steps." % max_steps, ConvergenceWarning)
<lifelines.CoxPHFitter: fitted with 115 observations, 19 censored>回答 2
Data Science用户
回答已采纳
发布于 2019-07-31 17:15:41
该软件包附带了关于CPH算法假设的优秀教程，甚至还为试验。比例危险。假设提供了一个函数。
如前所述，我首先遇到了一个收敛问题，但一旦解决了这个问题，我就得到了这样一个信息:其中一个变量违反了假设：
The ``p_value_threshold`` is set at 0.05. Even under the null hypothesis of no violations, some
covariates will be below the threshold by chance. This is compounded when there are many covariates.
Similarly, when there are lots of observations, even minor deviances from the proportional hazard
assumption will be flagged.
With that in mind, it's best to use a combination of statistical tests and visual tests to determine
the most serious violations. Produce visual plots using ``check_assumptions(..., show_plots=True)``
and looking for non-constant lines. See link [A] below for a full example.
1. Variable 'hsa-miR-181a-5p' failed the non-proportional test: p-value is 0.0331.
Advice 1: the functional form of the variable 'hsa-miR-181a-5p' might be incorrect. That is,there may be non-linear terms missing. The proportional hazard test used is very sensitive to
incorrect functional forms. See documentation in link [D] below on how to specify a functional form.
Advice 2: try binning the variable 'hsa-miR-181a-5p' using pd.cut, and then specify it in
`strata=['hsa-miR-181a-5p', ...]` in the call in `.fit`. See documentation in link [B] below.
Advice 3: try adding an interaction term with your time variable. See documentation in link [C]
below.
---
[A] https://lifelines.readthedocs.io/en/latest/jupyter_notebooks/Proportional%20hazard%20assumption.html
[B] https://lifelines.readthedocs.io/en/latest/jupyter_notebooks/Proportional%20hazard%20assumption.html#Bin-variable-and-stratify-on-it
[C] https://lifelines.readthedocs.io/en/latest/jupyter_notebooks/Proportional%20hazard%20assumption.html#Introduce-time-varying-covariates
[D] https://lifelines.readthedocs.io/en/latest/jupyter_notebooks/Proportional%20hazard%20assumption.html#Modify-the-functional-form
[E] https://lifelines.readthedocs.io/en/latest/jupyter_notebooks/Proportional%20hazard%20assumption.html#Stratification下面是我解决问题的步骤：
- 尽管完整的错误消息声明我有一个时变变量，但我100%没有任何时变变量。特别是，这个变量(错误消息声称它违反了假设)是每t的一个固定数。经过深入阅读，这可能会发生！另一种选择是直接纠正该变量/S；我设计了违反危险假设的具体变量为非线性(二次型)。此外，我已经将所有变量缩放到相同的范围。
- 我将其中一个变量(年龄)移动为
"strata"- w/o，这个问题还没有完全解决。请注意，我将"age“定义为几十年，因此它现在是一个带有9基数(相对较低)的分类变量。此外，我们知道年龄对“事件”(死亡时间)有影响，因为年龄与生存时间呈负相关。 - 我确实有严重的过度拟合，因为n<p在我的数据。我删除了相关变量，并添加了正则化项(L2惩罚)--完成了这项工作。
Data Science用户
发布于 2019-07-30 18:38:02
你好，这里是生命线作者。让我试着帮忙。
1)在fit开始运行时，您看到了任何Python警告吗？
2)我注意到你有115个观察，但超过190个变量。很可能系统被高估了:没有唯一的解决方案，您的模型将完全适应数据(更多的证据:一致性~= 1.0)。由于您的系数看起来非常小，您可能需要一个非常高的penalizer来“修复”这个问题，但是真正的解决方案是获取更多的数据。
3) alpha在合适的情况下不会改变任何东西，它只是用于置信区间。所以玩这个没什么意义。
( 4)从几何角度看，这个点接近最小值，但在那里很平，所以这个点正在发生很大的跳跃。这就是为什么三角洲仍然那么大的原因。同样，这是许多变量的结果，但数据点不多。
页面原文内容由Data Science提供。腾讯云小微IT领域专用引擎提供翻译支持
原文链接：
https://datascience.stackexchange.com/questions/56608
复制相关文章
相似问题
腾讯云开发者
Copyright © 2013 - 2026 Tencent Cloud. All Rights Reserved. 腾讯云 版权所有
深圳市腾讯计算机系统有限公司 ICP备案/许可证号：粤B2-20090059 
粤公网安备44030502008569号
腾讯云计算（北京）有限责任公司 京ICP证150476号 | 京ICP备11018762号