* Chunk tinygrad pkl below GitHub max size
* pull that out
* rm glob
* make work
* Single name def
* unused comment
* more cleanups
* revert that
* 10MB overhead
---------
Co-authored-by: Adeeb Shihadeh <adeebshihadeh@gmail.com>
* Revert "Revert tgwarp again (#37161)"
This reverts commit 45099e7fcd.
* Weird uv sizes
* Fix interleaving
* Fix on CPU
* make CPU safe
* Prevent corruption without clone
* Claude knows speeed
* fix interleaving
* less kernels
* blob caching
* This is still slightly faster
* Comment for blob cache
Hi! The point of this pr is to make the model run easier to read. On the latest tinygrad numpy().flatten() empirically does the same thing as the internal contiguous().realize().uop.base.buffer.numpy(). numpy() is also documented (docstrings), which can assist new contributors in learning what each potential execution does. Torq_boi or yassine, I know you want proof in the code base, so here it is. As of tinygrad commit 2f55005:
in tinygrad_repo/tinygrad/tensor.py
Lines 316-318 (def _buffer): ensure the tenso is contiguous() and realized() before accessing the raw buffer.
Line 378 (def numpy): Wraps the buffer access and adds a reshape to match the tensor shape.
self._buffer() is what executes contiguous().realize() and returns the buffer object.
Calling numpy() on that buffer object returns a 1D array (defined in tinygrad/device.py:193 via np.frombuffer).
The reshape(self.shape) at the end of Tensor.numpy() then adds dimensions to that 1D array. The added .flatten() removes those dimensions, flattening it back to a 1D array. Effectively the same as what is currently done, but less complex.
* Revert "revert tg calib and opencl cleanup (#37113)"
This reverts commit 51312afd3d.
* power draw is a lil higher
* just don't miss a cycle
* fix warp targets
* fix tinygrad dep
* Revert "Remove all the OpenCL (#37105)"
This reverts commit d5cbb89d84.
* Revert "rm common/mat.h"
This reverts commit 4ce701150a.
* Revert "Calibrate in tg (#36621)"
This reverts commit 593c3a0c8e.
* fix non-determinism in selfservice model build
also trim down model compile dependencies to the minimum required
* Apply suggestions from code review
---------
Co-authored-by: Shane Smiskol <shane@smiskol.com>
* Remove cython for transformations
* Add new test
* Switch back to program to fix mac builds
* Convert to Python instead
* Fix failing builds
* lint
* Implement conversion in pure python/numpy
* Add more tests
* Fix bugs in tests
* it's all common
* never getting fixed
* it's just tici
* reorders
* qcom2 -> tici
* Revert "qcom2 -> tici"
This reverts commit f4d849b2952cb0e662975805db6a1d32511ed392.
* Reapply "qcom2 -> tici"
This reverts commit 58b193cb8de872830f8a7821a339edca14e4a337.
* is tici
* lil more
* Revert "is tici"
This reverts commit a169be18d3fdcb3ef8317a63a89d8becadabfad8.
* Revert "Reapply "qcom2 -> tici""
This reverts commit 26f9c0e7d068fc8a1a5f07383b3616e619cd4e8c.
* qcom2 -> __tici__
* lil more
* mv lenv
* clean that up
* lil more]
* fix
* lil more
* Update Python packages
* not available anymore
* also this
* also this
* maybe?
* version
* try
* Revert "version"
This reverts commit 9ac4401b9ca59677b82736faff8baf66861df5f2.
* revert
* cffi
* issue
* comment
---------
Co-authored-by: Vehicle Researcher <user@comma.ai>
Co-authored-by: Maxime Desroches <desroches.maxime@gmail.com>
Harald Schäfer