Open Rheza001 opened 4 years ago
i only trained 80 data. should i create more?
i have a similar issue
After spending way too much time on this problem and trying endless solutions, I finally found what causes this issue. This error is particularly treacherous as in the most cases, it decides simply not to give an error message.
See the original thread here: https://github.com/weiliu89/caffe/issues/669#issuecomment-339542120
Before compiling, you must edit the source code a little bit. Go to caffe/src/caffe/util/math_functions.cpp and in line 247, you find this function, which you should edit to look like this:
void caffe_rng_uniform(const int n, Dtype a, Dtype b, Dtype* r) {
CHECK_GE(n, 0);
CHECK(r);
if (a > b) {
Dtype c = a;
a = b;
b = c;
}
CHECK_LE(a, b);
boost::uniform_real<Dtype> random_distribution(a, caffe_nextafter<Dtype>(b));
boost::variate_generator<caffe::rng_t*, boost::uniform_real<Dtype> >
variate_generator(caffe_rng(), random_distribution);
for (int i = 0; i < n; ++i) {
r[i] = variate_generator();
}
}
Note that I just added an if statement (that switches the variables a and b if a is larger than b) and removed the const flag in the parameter's line from Dtype a and Dtype b. Then simply do:
make clean
make -j$(nproc)
make py -j$(nproc)
make test -j$(nproc)
make runtest -j$(nproc) # You should run the tests after compiling to make sure you don't run into any other unexpected error.
For me, this worked very well!
Please help me. i'm stuck in this problem. when i tried to run ./train.sh, i got this error: