huggingface/candle
1
0
Fork 0
mirror of https://github.com/huggingface/candle.git synced 2025-06-16 18:48:51 +00:00
Code Issues Packages Projects Releases Wiki Activity

Validate the kernel size in pooling ops. (#1473)

Browse Source 
* Validate the kernel size in pooling ops.

* Revert the changes to basics.
This commit is contained in:
Laurent Mazare
2023-12-23 11:19:22 +01:00
committed by GitHub
parent 78d982e1bd
commit ba1fae590e
1 changed files with 16 additions and 12 deletions
Download Patch File Download Diff File Expand all files Collapse all files

28
candle-core/src/tensor.rs
View File

@ -396,7 +396,7 @@ impl Tensor {
device: &Device, device: &Device,
) -> Result<Self> { ) -> Result<Self> {
if D::is_zero(&step) { if D::is_zero(&step) {
crate::bail!("step cannot be zero") bail!("step cannot be zero")
} }
let mut data = vec![]; let mut data = vec![];
let mut current = start; let mut current = start;
@ -1041,6 +1041,9 @@ impl Tensor {
let kernel_size = kernel_size.to_usize2(); let kernel_size = kernel_size.to_usize2();
let stride = stride.to_usize2(); let stride = stride.to_usize2();
let (n, c, h, w) = self.dims4()?; let (n, c, h, w) = self.dims4()?;
if h < kernel_size.0 || w < kernel_size.1 {
bail!("kernel-size {kernel_size:?} is larger than the input size {h},{w}")
}
// https://pytorch.org/docs/stable/generated/torch.nn.AvgPool2d.html#torch.nn.AvgPool2d // https://pytorch.org/docs/stable/generated/torch.nn.AvgPool2d.html#torch.nn.AvgPool2d
let h_out = (h - kernel_size.0) / stride.0 + 1; let h_out = (h - kernel_size.0) / stride.0 + 1;
let w_out = (w - kernel_size.1) / stride.1 + 1; let w_out = (w - kernel_size.1) / stride.1 + 1;
@ -1076,6 +1079,9 @@ impl Tensor {
let kernel_size = kernel_size.to_usize2(); let kernel_size = kernel_size.to_usize2();
let stride = stride.to_usize2(); let stride = stride.to_usize2();
let (n, c, h, w) = self.dims4()?; let (n, c, h, w) = self.dims4()?;
if h < kernel_size.0 || w < kernel_size.1 {
bail!("kernel-size {kernel_size:?} is larger than the input size {h},{w}")
}
// https://pytorch.org/docs/stable/generated/torch.nn.MaxPool2d.html#torch.nn.MaxPool2d // https://pytorch.org/docs/stable/generated/torch.nn.MaxPool2d.html#torch.nn.MaxPool2d
let h_out = (h - kernel_size.0) / stride.0 + 1; let h_out = (h - kernel_size.0) / stride.0 + 1;
let w_out = (w - kernel_size.1) / stride.1 + 1; let w_out = (w - kernel_size.1) / stride.1 + 1;
@ -1798,7 +1804,7 @@ impl Tensor {
let is_permutation = let is_permutation =
dims.len() == self.rank() && (0..dims.len()).all(|i| dims.contains(&i)); dims.len() == self.rank() && (0..dims.len()).all(|i| dims.contains(&i));
if !is_permutation { if !is_permutation {
crate::bail!( bail!(
"dimension mismatch in permute, tensor {:?}, dims: {:?}", "dimension mismatch in permute, tensor {:?}, dims: {:?}",
self.dims(), self.dims(),
dims dims
@ -2293,7 +2299,7 @@ impl Tensor {
if left == 0 && right == 0 { if left == 0 && right == 0 {
Ok(self.clone()) Ok(self.clone())
} else if self.elem_count() == 0 { } else if self.elem_count() == 0 {
crate::bail!("cannot use pad_with_same on an empty tensor") bail!("cannot use pad_with_same on an empty tensor")
} else if left == 0 { } else if left == 0 {
let dim = dim.to_index(self.shape(), "pad_with_same")?; let dim = dim.to_index(self.shape(), "pad_with_same")?;
let r = self.narrow(dim, self.dim(dim)? - 1, 1)?; let r = self.narrow(dim, self.dim(dim)? - 1, 1)?;
@ -2457,13 +2463,13 @@ impl Tensor {
pub fn normalize_axis(&self, axis: i64) -> Result<usize> { pub fn normalize_axis(&self, axis: i64) -> Result<usize> {
let rank = self.rank() as i64; let rank = self.rank() as i64;
if rank <= axis { if rank <= axis {
crate::bail!("axis {axis} is too large, tensor rank {rank}") bail!("axis {axis} is too large, tensor rank {rank}")
} else if 0 <= axis { } else if 0 <= axis {
Ok(axis as usize) Ok(axis as usize)
} else { } else {
let naxis = rank + axis; let naxis = rank + axis;
if naxis < 0 { if naxis < 0 {
crate::bail!("axis {axis} is too small, tensor rank {rank}") bail!("axis {axis} is too small, tensor rank {rank}")
} }
Ok(naxis as usize) Ok(naxis as usize)
} }
@ -2525,14 +2531,14 @@ impl Tensor {
let src_dims = src.dims(); let src_dims = src.dims();
let self_dims = self.dims(); let self_dims = self.dims();
if self_dims.len() != src_dims.len() { if self_dims.len() != src_dims.len() {
crate::bail!( bail!(
"slice-assign requires input with the same rank {} <> {}", "slice-assign requires input with the same rank {} <> {}",
self_dims.len(), self_dims.len(),
src_dims.len() src_dims.len()
) )
} }
if self_dims.len() != ranges.len() { if self_dims.len() != ranges.len() {
crate::bail!( bail!(
"slice-assign requires input with the same rank as there are ranges {} <> {}", "slice-assign requires input with the same rank as there are ranges {} <> {}",
self_dims.len(), self_dims.len(),
ranges.len() ranges.len()
@ -2552,18 +2558,16 @@ impl Tensor {
std::ops::Bound::Excluded(v) => *v, std::ops::Bound::Excluded(v) => *v,
}; };
if end_excluded <= start_included { if end_excluded <= start_included {
crate::bail!( bail!("slice-assign: empty range for dim {i}, {start_included} {end_excluded}")
"slice-assign: empty range for dim {i}, {start_included} {end_excluded}"
)
} }
if self_dims[i] < end_excluded { if self_dims[i] < end_excluded {
crate::bail!( bail!(
"slice-assign: upper bound is out of range for dim {i}, {end_excluded} {}", "slice-assign: upper bound is out of range for dim {i}, {end_excluded} {}",
self_dims[i] self_dims[i]
) )
} }
if end_excluded - start_included != src_dims[i] { if end_excluded - start_included != src_dims[i] {
crate::bail!( bail!(
"slice-assign: the range for dim {i} ({start_included}..{end_excluded}) does not match the size of src {}", src_dims[i] "slice-assign: the range for dim {i} ({start_included}..{end_excluded}) does not match the size of src {}", src_dims[i]
) )
} }