Skip to content

Commit 078a2d9

Browse files
AleksMatDeepMind
authored andcommitted
Replace unicode escaped characters in ipynb files
PiperOrigin-RevId: 854213770
1 parent 8be970a commit 078a2d9

File tree

1 file changed

+2
-2
lines changed

1 file changed

+2
-2
lines changed

enn/colabs/epinet_demo.ipynb

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -25,7 +25,7 @@
2525
"For more details about Epinet, refer to the paper \n",
2626
"[Epistemic Neural Networks](https://arxiv.org/abs/2107.08924) (Osband et al., 2022).\n",
2727
"\n",
28-
"It's recommended to use `Runtime-\u003eChange Runtime Type` to pick a GPU for speed."
28+
"It's recommended to use `Runtime->Change Runtime Type` to pick a GPU for speed."
2929
]
3030
},
3131
{
@@ -303,7 +303,7 @@
303303
"id": "1sSq3q9ZXzy-"
304304
},
305305
"source": [
306-
"From the checkpoint, we can recover an enn sampler, which is a function that takes a batch of images and one random key, and returns multiple sample logits per input image. To recover the enn sampler for ResNet-50, we can use `make_enn_sampler_from_checkpoint` (from `enn/checkpoints/utils.py`) which takes the checkpoint and also the number of sample logits we want per image (`num_enn_samples`). Here we set `num_enn_samples=1`, as having `num_enn_samples \u003e 1` just results in multiple similar sample logits per input image."
306+
"From the checkpoint, we can recover an enn sampler, which is a function that takes a batch of images and one random key, and returns multiple sample logits per input image. To recover the enn sampler for ResNet-50, we can use `make_enn_sampler_from_checkpoint` (from `enn/checkpoints/utils.py`) which takes the checkpoint and also the number of sample logits we want per image (`num_enn_samples`). Here we set `num_enn_samples=1`, as having `num_enn_samples > 1` just results in multiple similar sample logits per input image."
307307
]
308308
},
309309
{

0 commit comments

Comments
 (0)