Xtra-Computing / thundersvm

ThunderSVM: A Fast SVM Library on GPUs and CPUs
Apache License 2.0
1.57k stars 218 forks source link

I build thundersvm followed the guide I can import SVC whitout error, but when I run the script and the process ended after training start INFO immediately #279

Open root-wang opened 6 months ago

root-wang commented 6 months ago

GPU info :NVIDIA-SMI 535.171.04 Driver Version: 535.171.04 CUDA Version: 12.2 NVIDIA GeForce RTX 3090

(base) root-wang@yao:~/machine-learn/frequenceFinger$  /usr/bin/env /opt/miniconda/envs/rw_torch/bin/python /home/root-wang/.vscode-server/extensions/ms-python.debugpy-2024.6.0-linux-x64/bundled/libs/debugpy/adapter/../../debugpy/launcher 51551 -- /home/root-wang/machine-learn/frequenceFinger/model/ML/svm.py 
2024-05-13 18:02:48,635 INFO [default] #instances = 150, #features = 4
2024-05-13 18:02:48,635 INFO [default] #classes = 3
2024-05-13 18:02:48,636 INFO [default] total memory size is 7.34255e-06 max mem size is 8
2024-05-13 18:02:48,636 INFO [default] free mem is 7.99999
2024-05-13 18:02:48,636 INFO [default] working set size = 64
2024-05-13 18:02:48,636 INFO [default] training start
(base) root-wang@yao:~/machine-learn/frequenceFinger$

I build thundersvm followed the guide I can import SVC whitout error, but when I run the script and the process ended after training start INFO immediately

from tabnanny import verbose
from torch import nn, Tensor
import torch

from sklearn.datasets import load_svmlight_file
from thundersvm import SVC
import torch

x, y = load_svmlight_file(
    "/home/root-wang/machine-learn/frequenceFinger/model/ML/test_dataset.txt"
)
clf = SVC(verbose=True, gamma=0.5, C=100)
clf.fit(x, y)

x2, y2 = load_svmlight_file(
    "/home/root-wang/machine-learn/frequenceFinger/model/ML/test_dataset.txt"
)
y_predict = clf.predict(x2)
score = clf.score(x2, y2)
clf.save_to_file("./model")

print("test score is ", score)