HYZhao

How do I set the random seed to reproduce the same result each time? I set it up as follows, but it doesn't work ``` tf.set_random_seed(1) np.random.seed(1) random.seed(1) ``` I'm...

[QST] Any experience to help me optimize on a particular problem size(a small M)?

5

I am a beginner to cutlass and I have reviewed many related documents and examples；I also have a general understanding of ThreadBlockShape, WarpShape, InstructionShape, NumStages; In my current problem domain,...

nccl not support for float16?

1

### Description ``` inputs = flow.ones([4, 10],dtype=flow.float32, placement=placement, sbp=flow.sbp.broadcast) weight = flow.ones([10, 4],dtype=flow.float32, placement=placement, sbp=flow.sbp.split(1)) * 2 y = flow.matmul(inputs, weight, ) y.to_global(placement, flow.sbp.broadcast).to_local() ``` this is ok, but when...

[QST] Tile is not evenly divisible

15

**What is your question?** I'm doing matrix multiplication manually with cute ``` Tensor A = make_tensor(make_gmem_ptr(Aptr), make_shape(m, k), make_stride(k, Int{})); Tensor B = make_tensor(make_gmem_ptr(Bptr), make_shape(n, k), make_stride(k, Int{})); Tensor C...

When I do fineweb-edu for classifier scoring, how do I overlap the tokenizer with the process of model infer?

3

I find that when tokenize, gpu utilization is always zero