microsoft microxcaling issues

microsoft / microxcaling

PyTorch emulation library for Microscaling (MX)-compatible data formats

MIT License

123 stars 14 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Update elemwise_ops.py

#24 jkshin94 opened 1 month ago
0
How is the matmul for MX format implemented?

#23 xijiu9 opened 1 month ago
0
Consider a PyPI package?

#22 awf opened 1 month ago
1
BFloat16 compatibility contributions

#21 hmellor opened 3 months ago
0
Add no-dim version of simd_reduce_sum/mean

#20 rizhao-msft closed 4 months ago
0
Add approximate kwarg to gelu to match pytorch iface

#19 rizhao-msft closed 4 months ago
0
Inference Error with OPT models

#18 ruisizhang123 closed 4 months ago
0
TypeError: gelu() got an unexpected keyword argument 'approximate'

#17 Arthur-Ling closed 4 months ago
4
Could You Offer a More Comprehensive Environment Setup Guide?

#16 Mr-Philo opened 5 months ago
0
Can MX formats emulate Block Floating Point (BFP)?

#15 JeevanBhoot closed 5 months ago
1
Error from test_conv.py and convolution.py to the latest version

#14 zks878 closed 5 months ago
1
Add the required changes to support mx auto injection.

#13 mgolub2-ms closed 6 months ago
0
Support for LSTMCell

#12 rjfnobre closed 6 months ago
2
Custom CUDA code vs. Pytorch CPU/GPU

#11 rjfnobre closed 6 months ago
1
Example ffn hangs on multi-gpu scenario (works on single gpu)

#10 lessw2020 closed 7 months ago
1
examples ffn_mx.py - Module not found error - no module named mx (needs path.append('..'))

#9 lessw2020 opened 7 months ago
2
Activations unit tests (gelu, tanh) hang on both A10 and A100 gpus...

#8 lessw2020 closed 7 months ago
23
Add MX integration guide + examples

#7 rizhao-msft closed 7 months ago
0
using fp32 data format to simulate mx data format is not equivalent？

#6 jackiechen19940708 opened 7 months ago
1
example or docs for getting started and converting an existing model to MX dtypes ala FP6?

#5 lessw2020 closed 7 months ago
4
957 is quantized as 896

#4 zhuango closed 8 months ago
1
MXINT8 elements: MX spec: 2's complement vs. README: sign-magnitude

#3 grayresearch closed 8 months ago
1
Action required: migrate or opt-out of migration to GitHub inside Microsoft

#2 microsoft-github-policy-service[bot] closed 8 months ago
3
Initial commit

#1 rizhao-msft closed 9 months ago
0