srush/annotated-s4

Segmentation fault!

Opened this issue · 0 comments

Hi! Great job!
Can you help me with this issue?
(ann1_s4) wsl@DESKTOP-CJL0ECN:/mnt/h/muhiddin/annotated-s4$ python -m s4.train dataset=quickdraw layer=s4 train.epochs=10 train.bsz=128 model.d_model=128 model.layer.N=64
2024-05-21 13:17:39.628139: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT
dataset: quickdraw
layer: s4
seed: 0
model:
d_model: 128
n_layers: 4
dropout: 0.0
prenorm: true
embedding: false
layer:
'N': 64
train:
epochs: 10
bsz: 128
lr: 0.001
lr_schedule: false
weight_decay: 0.01
checkpoint: false
suffix: null
sample: null
wandb:
mode: disabled
project: s4
entity: null

[] Warning: models are not being checkpoint
[
] Setting Randomness...
Segmentation fault

Whatever training code I run gives me this issue. I would appreciate any help!