* add caching for apt packages
* remove 'inputs' from apt cache key, use outputs instead of env
* remove unnecessary mkdir for partial
---------
Co-authored-by: George Hotz <[email protected]>
instead of running test with 3.10, add onnx to mypy which would have caught StrEnum regression. Several type annotation failed mypy now that does not affect running the code and were skipped for now
* start
* remove onnx.load from compile4 and move np to dropout
* clean up and enable test
* clean up
* move WebGPU ONNX test into MacOS (WebGPU)
* leave test in ONNX (CPU)
* fix raw_data init None, and simplify onnx_runner test a little?
* THESE TESTS ARE SO UGLY UGHH
* need to really think about how to structure the test
* wow LLMs are quite something
* not always on disk now
* also add external data loading test
* cleaner tests
* minimize diff and add const folding tests
* add external data loading too
* whoops add webgpu back.. but why was it not needed in the first place?
* better comment
* move webgpu test to macos(webgpu)?
* llm english so much better than me wow
* trigger CI to check flakiness
---------
Co-authored-by: chenyu <[email protected]>
* cast after sum
* comment out skipif
* minor fix
* only test IMAGE
* IMAGE is supported now
* simpler
* simplerr
* only cast if dtype is None
* dont need to change base_imaeg_type
* only cast when dtype is half
* add explicit test
* actually no, workflow seems better
* actually, keep both
* move test
* fix indent
---------
Co-authored-by: Utkarsh Gill <[email protected]>
Example: https://github.com/tinygrad/tinygrad/actions/runs/16215577171/job/45784110543?pr=11177#step:7:20
Caused by footguny exception in how `set -e` works:
```bash
python -m mypy --strict-equality --lineprecision-report . && cat lineprecision.txt
```
Will fail (and have non-zero exit code if run in interactive mode) but
because there is `&&` it won't count as script-terminating failure in a
script with `set -e` and instead as a test (similar to how fail of a
command in if condition won't count as a script-terminating failure
despite having non-zero exit code)
* start LLM app, tons of clean up required. target is 200 line ollama
* kind of works
* simpler
* add k/v cache
* with SYM=1, it loops
* no rope cache
* simpler
* more cleanups
* cleanups
* works
* argparse and comments
* from gguf
* generate is a function
* no copy from cpu
* fix max context pass in
* test
* improve test
* ai2_arc
* fix 8B, use less ram
* 136 lines
* WebGPU on Windows
* Fix dawn-python install
* New test
* pydeps
* Minor fix
* Only install dawn-python on windows webgpu
---------
Co-authored-by: George Hotz <[email protected]>
* kernel.py no longer permutes reduce axis [pr]
* delete tests that handcode uops
* regen of sops is broken...
* put import back
* just remove that
* disable those tests
* fix extract_dataset + tests
* add CI
* sops.gz itself is same as master
* yml + gzip -c + ge
* don't commit that
* bump limit to 1000
* axis=7
* test_tiny
* Don't use numpy inside hlb_cifar10 training loop
* Lint it
* jit it
* Drop the last half-batch
* Use gather for random_crop and reuse perms
* Wrap train_cifar in FUSE_ARANGE context
* No need to pass FUSE_ARANGE=1 to hlb_cifar10.py
* Add cutmix to jittable augmentations
* Remove .contiguous() from fetch_batches
* Fix indexing boundary
---------
Co-authored-by: Irwin1138 <[email protected]>
* change clang -march flag to -mcpu with fp16 disassembly test
* fix
* add capstone to macos dependencies
* just check no cast in test
* rm import
* woops
* lets check
* move check
* llvm init before cpu chcek
* try this
* bump autogen llvm version
* also update libclang?
* revert
* add comment
* skip llvm test and add comment
* linter
* fix
* add early verbose demo test
* is this how to write tests :s
* is definition drift even a thing? gemini says it is
* clean up
* better
* even better
* try add to CI
* doesn't work quite yet
* much more work to be done
* whoops
* partition the test heh
* skipif
* some nits for better names
* add webgpu test for onnxrunner
* fix reference links
* flush for now
geohot