 Guillaume Wenzek
|
4aee25223a
remove duplicated code
|
2 years ago |
|
Guillaume Wenzek
|
a771adc782
nicer ctypes_utils
|
2 years ago |
|
Guillaume Wenzek
|
ac4a2eada9
job.num_threads
|
2 years ago |
|
Guillaume Wenzek
|
f6d810543d
fix incremental decoding !
|
2 years ago |
|
Guillaume Wenzek
|
51bc5af890
GGML bug in ggml_set_2d_inplace
|
2 years ago |
|
Guillaume Wenzek
|
522b97234e
WIP: simple failing test case
|
2 years ago |
|
Guillaume Wenzek
|
2c543185e2
add kv_cache to fairseq2_model
|
2 years ago |
|
Guillaume Wenzek
|
f215d99664
drop kaldi tests
|
2 years ago |
|
Guillaume Wenzek
|
44a4ca129a
read model config from "layer_config"
|
2 years ago |
|
Guillaume Wenzek
|
cc23e2b1c7
prepare tests for flash_attn
|
2 years ago |
|
Guillaume Wenzek
|
18c919b0b8
update flash_attn tests
|
2 years ago |
|
Guillaume Wenzek
|
eb7810b81f
force little-endian
|
2 years ago |
|
Guillaume Wenzek
|
c31926c1a8
working out new way of saving hparams
|
2 years ago |
 Ning
|
d9061b89b3
Update test_unity_cpp.py
|
2 years ago |
|
Ning
|
1589c529bf
unity.cpp speech_encoder_frontend+speech_encoder+adaptor (#81)
|
2 years ago |
|
Ning
|
97283c1239
Self attention (SDPA) in unity.cpp (#34)
|
2 years ago |
|
Guillaume Wenzek
|
e38442d1f0
rename mask to attn_mask in MultiheadAttention_forward
|
2 years ago |
|
Guillaume Wenzek
|
6d23b4d1c2
ignore test file
|
2 years ago |
|
Guillaume Wenzek
|
6fbb465f2b
generate_sequence return full results
|
2 years ago |
|
Guillaume Wenzek
|
1756897d23
simplify _finalize_hypothesis
|
2 years ago |
|
Guillaume Wenzek
|
f49763de86
fix beam size and scores
|
2 years ago |
|
Guillaume Wenzek
|
184c1e1da3
};
|
2 years ago |
|
Guillaume Wenzek
|
c28db8c8ac
wip beam-size=2
|
2 years ago |
|
Guillaume Wenzek
|
3d69d4975f
batching -> StandardTransformerDecoder
|
2 years ago |
|
Guillaume Wenzek
|
b24dbe3030
batching -> TransformerEmbeddingFrontend_forward
|
2 years ago |
|
Guillaume Wenzek
|
b3e6d3c0c7
test_flatten
|
2 years ago |
|
Guillaume Wenzek
|
86993cbd00
fix StandardTransformerEncoder
|
2 years ago |
|
Guillaume Wenzek
|
28ed039370
fix MultiheadAttention_forward
|
2 years ago |
|
Guillaume Wenzek
|
b7b31a3978
rm python 3.11 idiom
|
2 years ago |
|
Guillaume Wenzek
|
81cdf80eb9
WIP: MultiheadAttention_forward
|
2 years ago |
