4373534d59
Metal: i64 basic support ( #1495 )
...
* Adds basic metal i64 support
* metal copy i64
2023-12-29 19:42:50 +01:00
f4a2787217
Merge pull request #1498 from huggingface/debugging_windows_ci
...
Fix CI
2023-12-29 12:33:50 +01:00
488e02a3f6
Merge pull request #1496 from bayedieng/unary
...
Implement urecip op for metal backend
2023-12-29 12:20:52 +01:00
adc95ca2bf
Ignore skipped.
2023-12-29 12:15:57 +01:00
4907c63ea1
Ignore stop on remote forks.
2023-12-29 12:12:10 +01:00
d76ac20e0e
Fix.
2023-12-29 12:06:38 +01:00
f5c98f22c7
Merge pull request #1491 from mimiquate/metal-errors
...
Improves metal's not implemented error messages
2023-12-29 12:03:40 +01:00
5b12fbb143
Trying to fix flakyness by making hub_2 and hub_3 serial tests (potential issue on mingw with mmap).
2023-12-29 11:13:33 +01:00
cc06ba2294
fix bad pattern matching and function name
2023-12-29 09:46:24 +00:00
a6bd0b47a5
Fix the CI.
2023-12-29 10:17:52 +01:00
b59b1b2bb6
remove generated png
2023-12-28 21:50:58 +00:00
3922b42c18
add urecip op to metal backend
2023-12-28 21:50:12 +00:00
1e442d4bb9
Fix lints for clippy 1.75. ( #1494 )
2023-12-28 20:26:20 +01:00
cd889c0f8a
add config_amazon_mistral_lite ( #1493 )
...
Co-authored-by: Ubuntu <danielclough@users.noreply.github.com>
2023-12-28 19:59:58 +01:00
8e93e76a91
fixes error message
2023-12-28 15:03:05 -03:00
b3e838f3e2
cargo fmt
2023-12-28 14:07:34 -03:00
8bf892403a
Improves metal's not implemented error messages
2023-12-28 11:04:06 -03:00
d35f0a1376
Bump the crate version to 0.3.3. ( #1490 )
2023-12-28 13:38:30 +01:00
65cb90bd40
Add some mention to SOLAR-10.7B in the readme. ( #1487 )
2023-12-27 15:25:39 +01:00
996a7f2e24
Rework the llama example config, add the solar model. ( #1485 )
2023-12-26 22:24:04 +01:00
3071ea6c3e
Use the new hub helper function. ( #1484 )
2023-12-26 09:44:30 +01:00
37c539f2b7
Helper function to load sharded safetensors files ( #1481 )
...
* Fix the quantized mistral example.
* Add a helper function to load sharded safetensors weights.
* Use the sharded loader.
2023-12-25 21:49:21 +01:00
eae3a20d43
Merge pull request #1479 from huggingface/upsample_metal
...
Adding upsample_nearest_2d.
2023-12-25 14:25:53 +01:00
13a5d15ebc
Adding upsample_nearest_2d.
2023-12-25 14:25:19 +01:00
1505d85276
Merge pull request #1461 from huggingface/metal-conv
...
Adding the convolutions (1d + 2d) to candle on metal.
2023-12-25 12:48:09 +01:00
95e18ef675
Fixing matmul for convolutions.
2023-12-25 12:29:34 +01:00
7135791dd5
Fix the quantized mistral example. ( #1478 )
2023-12-25 09:31:24 +01:00
88589d8815
Support mistral instruct v0.2. ( #1475 )
...
* Support mistral instruct v0.2.
* Use the safetensors model now that they are available.
2023-12-23 16:18:49 +01:00
5b35fd0fcf
MMLU evaluation for Phi. ( #1474 )
...
* MMLU evaluation for Phi.
* Improve the evaluation.
2023-12-23 15:28:36 +01:00
ba1fae590e
Validate the kernel size in pooling ops. ( #1473 )
...
* Validate the kernel size in pooling ops.
* Revert the changes to basics.
2023-12-23 11:19:22 +01:00
78d982e1bd
Fix for mamba 2.8b. ( #1472 )
2023-12-23 11:01:39 +01:00
d8b9a727fc
Support different mamba models. ( #1471 )
2023-12-23 10:46:02 +01:00
ceb78d3e28
Sketch the minimal mamba example. ( #1465 )
...
* Sketch the minimal mamba example.
* Fix rustfmt.
* Forward pass for mamba.
* Finish the forward pass.
* Inference fixes.
* Bugfixes.
* More fixes.
* Add a readme.
2023-12-22 00:28:50 +01:00
f6408a3779
feat: add clear_kv_cache to mistral and qmistral models ( #1464 )
2023-12-21 21:19:19 +01:00
10d94659c3
Adding the convolutions (1d + 2d) to candle on metal.
2023-12-21 10:39:24 +01:00
563a79afa1
make fn name generic ( #1459 )
...
Co-authored-by: Ubuntu <danielclough@users.noreply.github.com>
2023-12-21 02:16:31 +01:00
8ede5f4210
add fn config_chat_ml ( #1458 )
...
* add fn config_chat_ml
* Add a link to the original config.
---------
Co-authored-by: Ubuntu <danielclough@users.noreply.github.com>
Co-authored-by: laurent <laurent.mazare@gmail.com>
2023-12-20 21:03:24 +01:00
9fc210fae8
Merge pull request #1318 from huggingface/metal4
...
Starting to fix some tests.
2023-12-20 15:37:31 +01:00
9b5e4843a6
Optimizing decode matmul (Phi at 28tok/s on M3).
...
Adding some benchmark in order to help checking out matmul performance.
2023-12-20 09:54:19 +01:00
03641293ee
Clippy pass.
2023-12-18 15:22:43 +01:00
064ba17bd7
Remove print.
2023-12-18 11:04:16 +01:00
e8ee253ee0
Missing cast.
2023-12-18 11:01:18 +01:00
8bd3d6b94b
Index add.
2023-12-18 10:46:01 +01:00
6a3ca7da0c
Scatter add.
2023-12-18 10:32:22 +01:00
96f1a28e39
Add a simple full method. ( #1455 )
...
* Add a simple implementation of the full method.
* Add the docstring.
2023-12-17 20:15:57 -05:00
586b6f6fff
Adding gather op.
2023-12-17 23:34:12 +01:00
e4b0cc59f5
Adding CMP
2023-12-17 22:32:25 +01:00
0a6e0a8c9a
Implement randn (CPU-> device)
2023-12-17 19:09:08 +01:00
972903021c
Finish reduce kernels.
2023-12-17 19:07:00 +01:00
94817dac56
Bump the crate version to 0.3.2. ( #1452 )
2023-12-17 05:34:53 -06:00
Gonzalo