89a5b602a6
Move the conv1d layer to candle_nn. ( #117 )
2023-07-10 11:02:06 +01:00
b06e1a7e54
[nn] Move the Embedding and Activation parts. ( #116 )
...
* Share the Embedding and Activation parts.
* Tweak some activations.
2023-07-10 10:24:52 +01:00
9ce0f1c010
Sketch the candle-nn crate. ( #115 )
...
* Sketch the candle-nn crate.
* Tweak the cuda dependencies.
* More cuda tweaks.
2023-07-10 08:50:09 +01:00
ea5dfa69bc
Sketching the musicgen model. ( #66 )
...
* Skeleton files for musicgen.
* Add a musicgen model module.
* Sketch the model loading.
* Start adding the forward pass.
* More forward pass.
* Positional embeddings.
* Forward for the decoder layers.
* Add an empty function.
* Fix the musicgen weight names.
* More musicgen modeling.
* Add the T5 loading bits.
* Add the encodec config.
* Add the encodec module hierarchy.
* More Encodec modeling.
* Encodec modeling.
* Encodec modeling.
* Add more to the encodec modeling.
* Load the weights.
* Populate the resnet blocks.
* Also load the conv transpose weights.
* Split musicgen in multiple files.
2023-07-09 19:53:35 +01:00
c187f347bf
Make it easier to use whisper samples from the repo. ( #112 )
...
* Make it easier to use samples from the repo.
* Use f32 for accumulation in the f16/bf16 kernels.
2023-07-08 18:48:27 +01:00
f35cfc5e97
Sample with temperature. ( #106 )
2023-07-07 18:12:25 +01:00
03dffe9ecc
Use F32 for the reduce ops. ( #105 )
2023-07-07 17:55:21 +01:00
e923b3adc2
Add a KV cache to falcon. ( #104 )
2023-07-07 17:24:38 +01:00
05ff1cff66
Add some caching to the causal mask. ( #103 )
2023-07-07 12:56:44 +01:00
2df044f9a1
Clippy after rebase.
2023-07-07 09:22:09 +02:00
1ec221a749
Fixing falcon example.
2023-07-07 09:13:55 +02:00
d38a926c14
Convert the logits to f32 before extracting them. ( #102 )
2023-07-07 08:07:57 +01:00
bac4ef40f3
Add some text generation pipeline for falcon. ( #98 )
2023-07-07 06:34:22 +01:00
2b8e8c9f14
Bugfixes. ( #97 )
2023-07-06 23:26:11 +01:00
a3f3b93d16
Add the call to dense in the attention layer. ( #96 )
2023-07-06 23:22:08 +01:00
0a2c82e301
Merge pull request #92 from LaurentMazare/sync_hub
...
Creating new sync Api for `candle-hub`.
2023-07-07 00:10:47 +02:00
0f679fe42e
Fix some shape issues in falcon. ( #95 )
...
* Fix some shape issues.
* Use different dtypes.
2023-07-06 19:23:54 +01:00
4afa461b34
Sketch the Falcon model. ( #93 )
...
* Sketch the Falcon model.
* Add more substance to the falcon example.
* Falcon (wip).
* Falcon (wip again).
* Falcon inference.
* Get the weights from the api and properly generate the model.
* Use the proper model.
* Fix the file/revision names.
* Fix bias handling.
* Recompute the rot embeddings.
* Fix the input shape.
* Add the release-with-debug profile.
* Silly bugfix.
* More bugfixes.
* Stricter shape checking in matmul.
2023-07-06 19:01:21 +01:00
cae9212b70
Merge pull request #89 from LaurentMazare/extending_bert
...
Enabling `roberta` for the example (it's the same model as Bert, with just different naming.)
2023-07-06 16:29:26 +02:00
115629fe08
Creating new sync Api for candle-hub.
...
- `api::Api` -> `api::tokio::api` (And created new `api::sync::Api`).
- Remove `tokio` from all our examples.
- Using similar codebase for now instead of ureq (for simplicity).
2023-07-06 15:15:25 +02:00
3f291bdf9d
Enabling roberta for the example (it's the same model as Bert, with
...
just different naming.)
2023-07-06 13:25:21 +02:00
dd60bd84bb
MKL adjustments. ( #87 )
2023-07-06 11:37:27 +01:00
c297a50960
Add mkl support for matrix multiply. ( #86 )
...
* Fix some rebase issues.
* Use mkl instead.
* Use mkl in bert.
* Add the optional mkl feature.
* Conditional compilation based on the mkl feature.
* Add more mkl support.
2023-07-06 11:05:05 +01:00
cd230d26fe
Whisper tweaks ( #85 )
...
* Isolate the decoding bits of the whisper example.
* Decode -> Decoder.
* Add the suppress tokens filter.
* More suppress tokens.
2023-07-06 09:13:20 +01:00
d3418f1cff
Add the original whisper names as comment.
2023-07-06 07:57:03 +01:00
19ab5ea411
Merge pull request #78 from LaurentMazare/whisper_update
...
Adapting whisper for Hub use.
2023-07-06 07:21:58 +01:00
e2bfbcb79c
Support dim indexes in cat.
2023-07-05 20:39:08 +01:00
2c3d871b2e
Add a simpler way to specify the dim index for some ops.
2023-07-05 20:22:43 +01:00
174e57d216
Use avg pooling before the cosine similarity.
2023-07-05 17:05:50 +01:00
914e84deec
Add some sentence similarity comparision to the bert example.
2023-07-05 16:49:57 +01:00
653c5049f8
Adding auto download of audio file.
2023-07-05 15:21:53 +00:00
e85573a4bd
Adapting whisper for Hub use.
2023-07-05 14:35:27 +00:00
bae6d07b7e
Fix the position embeddings size.
2023-07-05 13:43:34 +01:00
93896f6596
Merge branch 'main' into upgrade_bert
2023-07-05 13:06:33 +01:00
d560855c2a
Bugfix for the mel filters.
2023-07-05 12:56:04 +01:00
63e5a266bf
Put everything together.
2023-07-05 12:19:21 +01:00
95f378ebb4
Read wav files.
2023-07-05 11:53:58 +01:00
26d1a7803f
Load the mel filters.
2023-07-05 11:20:33 +01:00
c701ee33a7
Add the mel filters.
2023-07-05 11:05:08 +01:00
648d1511d5
PCM conversion.
2023-07-05 11:02:49 +01:00
dd1d55f5c7
Mel spectogram computation.
2023-07-05 10:49:37 +01:00
f4c8a196a8
Mel spectogram.
2023-07-05 10:14:20 +01:00
7a6bc6d2dc
Mel spectogram computation (fft bits).
2023-07-05 09:54:12 +01:00
a824c5c3e3
Populate the no-speech probability.
2023-07-05 08:54:04 +01:00
d8f75ceeaa
Some polish.
2023-07-05 07:41:14 +00:00
9694e35db0
Clean the decode loop of the whisper example.
2023-07-05 08:37:26 +01:00
963c75cb89
Adding offline mode.
2023-07-05 07:19:57 +00:00
3ba4bfc501
More pretty printing.
2023-07-05 05:50:33 +01:00
8cf803d1a3
Split the model in a separate file.
2023-07-05 05:46:53 +01:00
9fe7a42895
More whisper sampling.
2023-07-04 22:18:07 +01:00
