Commit 9d0593b
fix: only cast float64 tensors in dict_to_device
Avoid converting boolean masks to float, which breaks ~ and torch.where.
Co-authored-by: Cursor <cursoragent@cursor.com>1 parent c6d5521 commit 9d0593b
1 file changed
Lines changed: 7 additions & 1 deletion
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
172 | 172 | | |
173 | 173 | | |
174 | 174 | | |
175 | | - | |
| 175 | + | |
| 176 | + | |
| 177 | + | |
| 178 | + | |
| 179 | + | |
| 180 | + | |
| 181 | + | |
176 | 182 | | |
177 | 183 | | |
178 | 184 | | |
| |||
0 commit comments