github/tinygrad
1
1
Fork 0
mirror of https://github.com/tinygrad/tinygrad.git synced 2026-01-09 15:08:02 -05:00
Code Issues Packages Projects Releases Wiki Activity

Refactor LoadOps (#910)

Browse Source 
* test

* work

* upd test

* loadops

* cleanups

* real ones

* remove LazyNumpyArray

* fix assign test

* remove range

* np.require

* llama uses arange kernels

* no caching consts

* fix enet

* torch load support

* tests cleanup

* fix shufflenet

* fix image

* fix torch_load test
This commit is contained in:
George Hotz
2023-06-03 09:40:43 -07:00
committed by GitHub
parent d58586bb17
commit 791530045d
20 changed files with 254 additions and 117 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
test/external/external_test_opt.py vendored
View File

@@ -27,7 +27,7 @@ class CLCache():
def __exit__(self, type, value, traceback):
print(f"cache: exiting with size {len(GlobalCounters.cache)}", f"allowed {self.allowed}" if self.allowed is not None else "")
if self.allowed is not None:
assert len(GlobalCounters.cache) <= self.allowed and (not self.strict or len(GlobalCounters.cache) == self.allowed), "used too many kernels!"
assert len(GlobalCounters.cache) <= self.allowed and (not self.strict or len(GlobalCounters.cache) == self.allowed), f"used too many kernels! {len(GlobalCounters.cache)} > {self.allowed}"
GlobalCounters.cache = None
from models.convnext import ConvNeXt
@@ -85,7 +85,7 @@ class TestInferenceMinKernels(unittest.TestCase):
args_tiny = {"dim": 512, "multiple_of": 256, "n_heads": 8, "n_layers": 4, "norm_eps": 1e-05, "vocab_size": 1000}
model = Transformer(**args_tiny)
for p in get_parameters(model): p.assign(np.zeros(p.shape, dtype=p.dtype.np))
with CLCache(82):
with CLCache(85):
model(Tensor([[1,2,3,4]]), 0).realize()
@unittest.skipUnless(Device.DEFAULT == "GPU", "Not Implemented")