andr3/gotch
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

fixed some old APIs at vision/aug

Browse Source
This commit is contained in:
sugarme 2022-02-05 16:58:15 +11:00
parent a4e5f38705
commit 04d738db1e
2 changed files with 5 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
CHANGELOG.md
View File

@ -10,6 +10,8 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
- added `nn.VarStore.Summary()` method
- fixed incorrect tensor method `ts.Meshgrid` -> `Meshgrid`
- added new API `ConstantPadNdWithVal` `ato_constant_pad_nd` with padding value.
- fixed "nn/rnn NewLSTM() clashed weight names"
- fixed some old API at `vision/aug/function.go`
## [Nofix]
- ctype `long` caused compiling error in MacOS as noted on [#44]. Not working on linux box.

6
vision/aug/function.go
View File

@ -16,7 +16,7 @@ func gaussianKernel1D(ks int64, sigma float64, dtype gotch.DType, device gotch.D
x := ts.MustLinspace(ts.IntScalar(-ksHalf), ts.IntScalar(ksHalf), []int64{ks}, dtype, device)
// pdf = torch.exp(-0.5 * (x / sigma).pow(2))
pdf := x.MustDivScalar(ts.FloatScalar(sigma), true).MustPow(ts.IntScalar(2), true).MustMulScalar(ts.FloatScalar(0.5), true).MustExp(true)
pdf := x.MustDivScalar(ts.FloatScalar(sigma), true).MustPowTensorScalar(ts.IntScalar(2), true).MustMulScalar(ts.FloatScalar(0.5), true).MustExp(true)
// kernel1d = pdf / pdf.sum()
pdfSum := pdf.MustSum(dtype, false)
kernel1d := pdf.MustDiv(pdfSum, true)
@ -1223,7 +1223,7 @@ func autocontrast(img *ts.Tensor) *ts.Tensor {
maxTsView.MustDrop()
// scale = bound / (maximum - minimum)
scale := maxTs.MustSub(minTs, false).MustPow(ts.IntScalar(-1), true).MustMulScalar(ts.FloatScalar(bound), true)
scale := maxTs.MustSub(minTs, false).MustPowTensorScalar(ts.IntScalar(-1), true).MustMulScalar(ts.FloatScalar(bound), true)
//
// return ((img - minimum) * scale).clamp(0, bound).to(img.dtype)
out := img.MustSub(minTs, false).MustMul(scale, true).MustClamp(ts.IntScalar(0), ts.FloatScalar(bound), true).MustTotype(dtype, true)
@ -1410,7 +1410,7 @@ func scaleChannel(imgChan *ts.Tensor) *ts.Tensor {
// lut = torch.div(torch.cumsum(hist, 0) + torch.div(step, 2, rounding_mode='floor'), step, rounding_mode='floor')
halfStep := step.MustFloorDivideScalar(ts.FloatScalar(2.0), false)
lut := histo.Must_Cumsum(0, true).MustAdd(halfStep, true).MustFloorDivide(step, true)
lut := histo.MustCumsum(0, histo.DType(), true).MustAdd(halfStep, true).MustFloorDivide(step, true)
step.MustDrop()
halfStep.MustDrop()