antmachineintelligence / mtgbmcode

mtgbmcode
MIT License
166 stars 42 forks source link

example1 run error #5

Open min13s opened 2 years ago

min13s commented 2 years ago

try to run the example1 in linux terminal delete the part concerning NN

for MT-GBM part, it runs successfully

[LightGBM] [Debug] Trained a tree with leaves = 10 and max_depth = 6 [LightGBM] [Warning] No further splits with positive gain, best gain: -inf [LightGBM] [Debug] Trained a tree with leaves = 13 and max_depth = 6 [LightGBM] [Warning] No further splits with positive gain, best gain: -inf [LightGBM] [Debug] Trained a tree with leaves = 13 and max_depth = 6 [LightGBM] [Warning] No further splits with positive gain, best gain: -inf [LightGBM] [Debug] Trained a tree with leaves = 13 and max_depth = 6 [LightGBM] [Warning] No further splits with positive gain, best gain: -inf [LightGBM] [Debug] Trained a tree with leaves = 13 and max_depth = 6 [200] training's rmse: 377.142 valid_1's rmse: 363.75 (942, 22) (31, 22) [LightGBM] [Warning] Unknown parameter: verboses [LightGBM] [Warning] Unknown parameter: verboses [LightGBM] [Info] Total Bins 3159 [LightGBM] [Info] Number of data points in the train set: 942, number of used features: 22 [LightGBM] [Warning] Unknown parameter: verboses [LightGBM] [Info] Start training from score 6507.648620 [LightGBM] [Warning] No further splits with positive gain, best gain: -inf Segmentation fault (core dumped)

but for the normal lightgbm, there is segment falt

zyg0121 commented 2 years ago

I have the same problem with you. Do you solve this problem now?

AI-amateur commented 2 years ago

I also have the same problem as above