zhangyiss/mlx
1
0
Fork 0
mirror of https://github.com/ml-explore/mlx.git synced 2025-08-12 04:06:39 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity

Add scatter_min VJP (#462)

Browse Source
This commit is contained in:
Tristan Bilot 2024-01-16 09:37:40 +01:00 committed by GitHub
parent 92a2fdd577
commit f44c132f4a
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
2 changed files with 31 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
mlx/primitives.cpp
View File

@ -2130,10 +2130,11 @@ std::vector<array> Scatter::vjp(
case Scatter::None: case Scatter::None:
case Scatter::Sum: case Scatter::Sum:
case Scatter::Max: case Scatter::Max:
case Scatter::Min:
break; break;
default: default:
throw std::runtime_error( throw std::runtime_error(
"[scatter] VJP implemented only for scatter and scatter_add"); "[scatter] VJP not implemented for scatter_prod");
} }
const array& values = primals[0]; const array& values = primals[0];
@ -2145,6 +2146,8 @@ std::vector<array> Scatter::vjp(
switch (reduce_type_) { switch (reduce_type_) {
case Scatter::Max: case Scatter::Max:
return scatter_max(values, indices, updates, axes_, stream()); return scatter_max(values, indices, updates, axes_, stream());
case Scatter::Min:
return scatter_min(values, indices, updates, axes_, stream());
default: default:
return array({}); return array({});
} }
@ -2169,7 +2172,8 @@ std::vector<array> Scatter::vjp(
// The input array values are kept so they all get gradients // The input array values are kept so they all get gradients
vjps.push_back(cotangents[0]); vjps.push_back(cotangents[0]);
break; break;
case Scatter::Max: { case Scatter::Max:
case Scatter::Min: {
auto mask = where(result == values, array({1}), array({0})); auto mask = where(result == values, array({1}), array({0}));
vjps.push_back(multiply(cotangents[0], mask)); vjps.push_back(multiply(cotangents[0], mask));
break; break;
@ -2191,7 +2195,8 @@ std::vector<array> Scatter::vjp(
gather(cotangents[0], indices, axes_, slice_sizes, stream())); gather(cotangents[0], indices, axes_, slice_sizes, stream()));
break; break;
} }
case Scatter::Max: { case Scatter::Max:
case Scatter::Min: {
auto slice_sizes = cotangents[0].shape(); auto slice_sizes = cotangents[0].shape();
for (auto ax : axes_) { for (auto ax : axes_) {
slice_sizes[ax] = 1; slice_sizes[ax] = 1;

23
python/tests/test_autograd.py
View File

@ -316,6 +316,29 @@ class TestAutograd(mlx_tests.MLXTestCase):
self.assertTrue(mx.allclose(vjps[0], mx.array([[4.0], [5.0], [6.0]]))) self.assertTrue(mx.allclose(vjps[0], mx.array([[4.0], [5.0], [6.0]])))
self.assertTrue(mx.allclose(vjps[1], mx.array([[[5.0]]]))) self.assertTrue(mx.allclose(vjps[1], mx.array([[[5.0]]])))
def test_scatter_min_vjp(self):
def fun(src, updates):
x = src.at[1].minimum(updates)
return x
cotan = mx.array([4.0, 5.0, 6.0])
_, vjps = mx.vjp(fun, [mx.array([1.0, 2.0, 3.0]), mx.array([[3.0]])], [cotan])
mx.eval(vjps)
# Update larger than value
self.assertTrue(mx.allclose(vjps[0], mx.array([4.0, 5.0, 6.0])))
self.assertTrue(mx.allclose(vjps[1], mx.array([0.0])))
cotan = mx.array([[4.0], [5.0], [6.0]])
_, vjps = mx.vjp(
fun, [mx.array([[1.0], [2.0], [3.0]]), mx.array([[[2.0]]])], [cotan]
)
mx.eval(vjps)
# Update and value are equal
self.assertTrue(mx.allclose(vjps[0], mx.array([[4.0], [5.0], [6.0]])))
self.assertTrue(mx.allclose(vjps[1], mx.array([[[5.0]]])))
def test_vjp_types(self): def test_vjp_types(self):
def fun(x): def fun(x):
return x return x