summaryrefslogtreecommitdiff
path: root/candle-examples
Commit message (Collapse)AuthorAgeFilesLines
* Support more mistral models. (#1927)Laurent Mazare2024-03-241-2/+39
| | | | | * Support more mistral models. * Use the appropriate rope parameter.
* Improve the error message on overlong prompts. (#1908)Laurent Mazare2024-03-211-0/+14
|
* whisper readme (#1899)Sanchit Gandhi2024-03-211-3/+4
|
* support distil-large-v3 (#1898)Sanchit Gandhi2024-03-211-1/+5
|
* Update the image crate + use the re-exported version. (#1893)Laurent Mazare2024-03-213-24/+25
| | | | | * Update the image crate + use the re-exported version. * Update to using ab_glyph.
* Avoid copying the data on squeeze and unsqueeze. (#1884)Laurent Mazare2024-03-201-3/+3
| | | | | | | | | * Avoid copying the data on squeeze and unsqueeze. * Fix the quantized llama example. * Unrelated fix for the quantized stable-lm example on cuda. * Fix for mamba on cuda (unrelated to the PR).
* Apply rustfmt. (#1873)Laurent Mazare2024-03-181-3/+3
|
* Add a DQN example to the reinforcement-learning section (#1872)Gabriel2024-03-182-0/+121
|
* Microphone support for the encodec example. (#1866)Laurent Mazare2024-03-182-7/+31
|
* Improve the encodec example: handle resampling. (#1865)Laurent Mazare2024-03-184-64/+309
| | | | | * Improve the encodec example: handle resampling. * Play the audio directly.
* Fix for the encodec example. (#1861)Laurent Mazare2024-03-171-2/+1
|
* Move the image tensor to the appropriate device. (#1856)Laurent Mazare2024-03-1610-10/+10
|
* StableLM-2 models were updated to use GPT-2 tokenization. (#1847)Jani Monoses2024-03-142-13/+1
|
* Update gemma README (#1843)Tyler Rockwood2024-03-131-1/+1
| | | | | * Update gemma README * Fixit
* Fix the model path for rwkv. (#1825)Laurent Mazare2024-03-091-1/+1
|
* Quantized version of the metavoice model. (#1824)Laurent Mazare2024-03-091-8/+36
| | | | | * Quantized version of the metavoice model. * Integrate the quantized version of metavoice.
* Fast CPU kernel for transposed 1d convolutions. (#1822)Laurent Mazare2024-03-081-1/+1
| | | | | * Fast CPU kernel for transposed 1d convolutions. * Bugfix.
* Add a --seed argument to the stable-diffusion example. (#1812)Niklas Hallqvist2024-03-081-0/+8
| | | | | | | | | * Add a --seed argument to the stable-diffusion example. * Make the case when no seed is specified, that it will not be set, but use the engine's default. This will make the CPU engine work again when no --seed is given, and will cause a bailout when a seed is there, as the engine does not currently support it. --------- Co-authored-by: niklas <niklas@appli.se>
* Add a flag to select the dtype used in metavoice. (#1805)Laurent Mazare2024-03-051-10/+22
|
* Add an initial Segformer implementation (#1617)Jiayu Liu2024-03-033-0/+935
| | | | | | | | | * add segformer * Make the id2label field optional. --------- Co-authored-by: laurent <laurent.mazare@gmail.com>
* More metavoice tweaks. (#1796)Laurent Mazare2024-03-031-1/+5
|
* Normalize loudness of the generated audio (#1795)Laurent Mazare2024-03-035-2/+543
| | | | | | | | | | | | | * Normalize loudness of the generated audio. * Lints. * One more lint. * Avoid running the bs1770 tests. * Another attempt at discarding doc comments. * Also normalize the loudness in the encodec example.
* Use the same padding in metavoice as in the python version. (#1794)Laurent Mazare2024-03-031-3/+6
|
* Tweaks to run metavoice on metal (#1792)Laurent Mazare2024-03-031-4/+11
| | | | | | | * Enable tanh + tweak conv-transpose. * Run the encodec decoding on cpu. * Clippy fixes.
* Metavoice position fix (#1791)Laurent Mazare2024-03-021-5/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | * Add the metavoice transformer. * Sketch the speaker-encoder module. * Adding to the metavoice model. * Start adding the metavoice example. * Get some logits out. * Load the second stage model. * Get the second step to run. * Tweak the example. * Add encodec tilting. * Glue the different bits together. * Fix a shape issue. * Use a constant. * BPE tokenization. * Fix the position index in metavoice.
* Add the instruction finetuned gemma variants. (#1790)Laurent Mazare2024-03-021-0/+2
|
* Metavoice - first cut (#1717)Laurent Mazare2024-03-022-0/+236
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | * Add the metavoice transformer. * Sketch the speaker-encoder module. * Adding to the metavoice model. * Start adding the metavoice example. * Get some logits out. * Load the second stage model. * Get the second step to run. * Tweak the example. * Add encodec tilting. * Glue the different bits together. * Fix a shape issue. * Use a constant. * BPE tokenization. * Add a warning.
* fix rwkv example eos token (#1785)Jack Shih2024-03-011-0/+5
|
* Mention rwkv v6 in the readmes. (#1784)Laurent Mazare2024-03-011-2/+2
|
* EfficientVit (MSRA) model (#1783)Jani Monoses2024-03-012-0/+119
| | | | | * Add EfficientVit (Microsoft Research Asia) model. * Mention models in README
* add models of rwkv v6 and quantized rwkv v6 (#1781)Jack Shih2024-03-011-15/+41
| | | | | * add models of rwkv v6 and quantized rwkv v6 * fix ci clippy fail
* Add the StarCoder2 model. (#1779)Laurent Mazare2024-02-282-1/+254
| | | | | | | * Add the StarCoder2 model. * Add the example code and get things to work. * And also tweak the readme.
* Add a flag to force running the quantized model on CPUs. (#1778)Laurent Mazare2024-02-281-1/+5
| | | | | * Add a flag to force running the quantized model on CPUs. * Add encodec to the readme.
* Support more modes in the encodec example. (#1777)Laurent Mazare2024-02-287-641/+159
| | | | | * Support more modes in the encodec example. * Remove the old encodec model from the musicgen bits.
* Make some dependencies optional in the examples. (#1776)Laurent Mazare2024-02-281-2/+14
|
* Encodec encoding demo. (#1775)Laurent Mazare2024-02-281-1/+16
|
* Encodec model. (#1771)Laurent Mazare2024-02-274-0/+114
| | | | | | | | | | | | | | | * Encodec model. * Fixes. * Add the padding functions. * Get the LSTM bit to work. * Get the encodec model to generate some tokens (decoder only for now). * Minor tweak. * Minor tweak.
* Add an option to split the prompt. (#1766)Laurent Mazare2024-02-271-1/+14
|
* add quantized rwkv v5 model (#1743)Jack Shih2024-02-251-4/+38
| | | | | | | | | * and quantized rwkv v5 model * Integrate the quantized rwkv model in the initial example. --------- Co-authored-by: laurent <laurent.mazare@gmail.com>
* Cuda acceleration for quantized model. (#1754)Laurent Mazare2024-02-251-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * Boilerplate for the quantized cuda support. * More basic cuda support. * More cuda quantization (quantize on cpu for now). * Add the dequantization bit. * Start adding some dedicated cuda kernels from llama.cpp. * Move the kernel code. * Start interfacing with the kernel. * Tweak the kernel launch params. * Bugfix for quantized metal. * Fix some clippy lints. * Tweak the launch parameters. * Tweak cuda basics to perform a quantized matmul. * Perform the dequantization on the cpu + use cublas for matmul. * Add the dequantization kernel. * Test the qmatmul. * More kernels. * Matmul-vec kernel. * Add a couple kernels. * More dequantization kernels.
* Fix the eos token for gemma. (#1753)Laurent Mazare2024-02-241-2/+2
|
* Fix typo in README (#1740)Daniel Varga2024-02-221-1/+1
|
* Make the cache for the llama model explicit too. (#1745)Laurent Mazare2024-02-221-3/+3
|
* Explicit caching in llama2.c.laurent2024-02-222-20/+21
|
* Add the Gemma models. (#1741)Laurent Mazare2024-02-212-0/+281
| | | | | | | | | | | | | | | | | | | | | * Add the Gemma models. * Add the gemma example. * Adapt the RmsNorm. * Get the 2b model to work. * 7b support. * Use the config head dim. * Yet another fix. * Make the matrixes contiguous. * Also get the 7b model to work. * And add to the readme.
* Use the tokenizer-output-stream in the llama example. (#1715)Laurent Mazare2024-02-154-20/+17
| | | | | * Use the tokenizer-output-stream in the llama example. * Also use tokenizer-output-stream for llama2-c.
* Add a readme for rwkv. (#1712)Laurent Mazare2024-02-141-0/+17
|
* Custom tokenizer for rwkv. (#1711)Laurent Mazare2024-02-141-38/+13
| | | | | | | * Custom tokenizer for rwkv. * Custom tokenizer. * Getting the tokenizer to work.
* Add the RWKV model (v5). (#1707)Laurent Mazare2024-02-141-0/+290
| | | | | | | | | | | | | | | | | | | | | | | * Start adding the RWKV model. * More of the forward step. * Handle rescaling. * FeedForward. * More work on RWKV. * Better state tracking. * Finish a first pass on forward. * Fix the shape mismatches. * Do not rescale in f32. * Rename to rwkv-v5. * Add the new models to the readme.
* Add ConvNeXt-V2 and smaller model variants. (#1709)Jani Monoses2024-02-142-15/+40
|
generated by cgit v1.2.3 (git 2.39.1) at 2025-04-13 12:39:10 +0000