Hugging Face's logo

Moon-Young-Choi
/
HUTACE-I5x64x64-B8x8-T65-0.83M

Reinforcement Learning
PyTorch
Safetensors
hutace
exploration
actor-critic
terrain
partial-observation
mapzen
Model card Files
xet
 Community
Moon-Young-Choi commited on
Commit
913c735
·
verified ·
1 Parent(s): 529e4ef

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +1 -1
README.md CHANGED
@@ -29,7 +29,7 @@ Input is `float32[5,64,64]` in this order:
29
  4. visit count normalized as `clip(count/3, 0, 1)`
30
  5. remaining-fuel ratio.
31
 
32
- The model returns 4,096 waypoint logits and one critic value. Apply the common local action mask before selecting a deterministic argmax.
33
  ![image](https://cdn-uploads.huggingface.co/production/uploads/6a5cc28a2257c24c62a73afa/QWgWw8ZctaPsqAYNKcRAa.png)
34
 
35
  ## Quick use
 
29
  4. visit count normalized as `clip(count/3, 0, 1)`
30
  5. remaining-fuel ratio.
31
 
32
+ The model returns 4,096 waypoint logits and one critic value.
33
  ![image](https://cdn-uploads.huggingface.co/production/uploads/6a5cc28a2257c24c62a73afa/QWgWw8ZctaPsqAYNKcRAa.png)
34
 
35
  ## Quick use