search

danijar / dreamerv3

Mastering Diverse Domains through World Models
https://danijar.com/dreamerv3
MIT License
1.28k stars 219 forks source link

Crafter-model size #98

Closed vaggelisT closed 5 months ago

vaggelisT commented 10 months ago

Hello,

I have changed the model size in example.py file from medium to large. config = config.update(dreamerv3.configs['large'])

but i get this error:

...... /envs/TestingEnv/lib/python3.8/site-packages/jax/_src/numpy/ufuncs.py:97 │ │ in fn │ │ │ │ 94 │ lax_doc: bool = False) -> BinOp: │ │ 95 def fn(x1, x2, /): │ │ 96 │ x1, x2 = promote_args(numpy_fn.name, x1, x2) │ │ ❱ 97 │ return laxfn(x1, x2) if x1.dtype != np.bool else bool_lax_fn(x1, x2) │ │ 98 fn.qualname = f"jax.numpy.{numpy_fn.name}" │ │ 99 fn = jit(fn, inline=True) │ │ 100 if lax_doc: │ ╰──────────────────────────────────────────────────────────────────────────────────────────────────╯ TypeError: add got incompatible shapes for broadcasting: (1, 1024), (1, 2048).

any suggestion?

AgiNetz commented 10 months ago

You most likely didn't set a different logging directory for the run and the model is trying to load an incompatible checkpoint automatically. I also spent some time on this before realizing it...