Doesn't converge, how to converge #565
Replies: 2 comments 1 reply
-
|
It is better to focus on hypothesis driven models where only some
parameters vary by condition - e.g. only in rare cases would t differ by
condition for example. Also there are known issues with convergence with t
fitted hierarchically, you can find solutions to try on other discussion
threads here. (others in the works)
M
…On Tue, Aug 20, 2024 at 4:14 PM Jingzhu Chen ***@***.***> wrote:
for a model like this:
model_0 = hssm.HSSM(data=data,
hierarchical=True,
prior_settings="safe",
include=[
{
"name": "v",
"formula": "v ~ condition + diff + (1|participant_id)",
},
{
"name": "a",
"formula": "a ~ condition + diff + (1|participant_id)",
},
{
"name": "t",
"formula": "t ~ condition + diff + (1|participant_id)",
},
],
)
Are these hyperparameters enough to make it converge?
infer_model_0 = model_0.sample(
sampler="nuts_numpyro", chains=4, cores=1, draws=1000, tune=2000,
target_accept=0.95
)
Sometimes I get error messages like this, that says target_accept has to
be higher, but how high is high enough, how many chains, cares, draws, tune
are enough for how big of a dataset? Any advice on how to set it up?
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
Compiling.. : 0%| | 0/3000 [00:00<?, ?it/s]
0%| | 0/3000 [00:00<?, ?it/s]
Compiling.. : 0%| | 0/3000 [00:00<?, ?it/s]
0%| | 0/3000 [00:00<?, ?it/s]
Compiling.. : 0%| | 0/3000 [00:00<?, ?it/s]
0%| | 0/3000 [00:00<?, ?it/s]
Compiling.. : 0%| | 0/3000 [00:00<?,
?it/s]C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
Running chain 3: 0%| | 0/3000 [00:09<?, ?it/s]
Running chain 0: 0%| | 0/3000 [00:09<?, ?it/s]
Running chain 1: 0%| | 0/3000 [00:09<?, ?it/s]
Running chain 3: 5%|███ | 150/3000 [07:45<2:24:14, 3.04s/it]
Running chain 1: 5%|███ | 150/3000 [08:27<2:37:34, 3.32s/it]
Running chain 0: 5%|███ | 150/3000 [10:53<3:23:53, 4.29s/it]
Running chain 3: 10%|██████ | 300/3000 [13:51<2:00:54, 2.69s/it]
Running chain 0: 10%|██████ | 300/3000 [17:53<2:33:35, 3.41s/it]
Running chain 1: 15%|█████████ | 450/3000 [21:10<1:53:13, 2.66s/it]
Running chain 0: 15%|█████████ | 450/3000 [25:47<2:20:09, 3.30s/it]
Running chain 0: 20%|████████████ | 600/3000 [32:32<2:02:25, 3.06s/it]
Running chain 2: 15%|█████████ | 450/3000 [34:05<3:10:18, 4.48s/it]
Running chain 1: 20%|████████████ | 600/3000 [36:07<2:38:51, 3.97s/it]
Running chain 0: 25%|███████████████ | 750/3000 [41:29<2:01:51, 3.25s/it]
Running chain 2: 20%|████████████ | 600/3000 [43:04<2:45:10, 4.13s/it]
Running chain 3: 25%|███████████████ | 750/3000 [44:13<2:03:22, 3.29s/it]
Running chain 1: 25%|███████████████ | 750/3000 [46:10<2:29:38, 3.99s/it]
Running chain 1: 30%|██████████████████ | 900/3000 [53:58<2:09:18,
3.69s/it]
Running chain 2: 25%|███████████████ | 750/3000 [57:48<2:58:38, 4.76s/it]
Running chain 0: 30%|█████████████████▍ | 900/3000 [1:02:30<2:55:04,
5.00s/it]
Running chain 2: 30%|█████████████████▍ | 900/3000 [1:06:52<2:33:12,
4.38s/it]
Running chain 0: 35%|███████████████████▉ | 1050/3000 [1:11:39<2:28:19,
4.56s/it]
Running chain 1: 45%|█████████████████████████▋ | 1350/3000
[1:18:01<1:33:06, 3.39s/it]
Running chain 3: 30%|█████████████████▍ | 900/3000 [1:24:22<4:27:07,
7.63s/it]
Running chain 1: 50%|████████████████████████████▌ | 1500/3000
[1:25:59<1:23:06, 3.32s/it]
Running chain 0: 40%|██████████████████████▊ | 1200/3000 [1:29:24<2:41:09,
5.37s/it]
Running chain 1: 55%|███████████████████████████████▎ | 1650/3000
[1:32:28<1:09:48, 3.10s/it]
Running chain 3: 35%|███████████████████▉ | 1050/3000 [1:35:09<3:32:44,
6.55s/it]
Running chain 0: 45%|█████████████████████████▋ | 1350/3000
[1:39:59<2:17:56, 5.02s/it]
Running chain 1: 60%|██████████████████████████████████▏ | 1800/3000
[1:44:49<1:13:11, 3.66s/it]
Running chain 0: 50%|████████████████████████████▌ | 1500/3000
[1:48:15<1:52:12, 4.49s/it]
Running chain 3: 40%|██████████████████████▊ | 1200/3000 [1:49:58<3:10:28,
6.35s/it]
Running chain 1: 65%|█████████████████████████████████████ | 1950/3000
[1:53:38<1:03:21, 3.62s/it]
Running chain 2: 50%|████████████████████████████▌ | 1500/3000
[1:56:49<1:51:53, 4.48s/it]
Running chain 1: 70%|█████████████████████████████████████████▎ |
2100/3000 [1:57:13<44:23, 2.96s/it]
Running chain 3: 45%|█████████████████████████▋ | 1350/3000
[2:00:49<2:37:19, 5.72s/it]
Running chain 1: 75%|████████████████████████████████████████████▎ |
2250/3000 [2:02:02<33:06, 2.65s/it]
Running chain 1: 80%|███████████████████████████████████████████████▏ |
2400/3000 [2:05:21<22:30, 2.25s/it]
Running chain 1: 85%|██████████████████████████████████████████████████▏ |
2550/3000 [2:08:28<14:37, 1.95s/it]
Running chain 2: 55%|███████████████████████████████▎ | 1650/3000
[2:10:12<1:46:44, 4.74s/it]
Running chain 1: 90%|█████████████████████████████████████████████████████
| 2700/3000 [2:11:23<08:34, 1.71s/it]
Running chain 0: 55%|███████████████████████████████▎ | 1650/3000
[2:15:23<2:24:47, 6.44s/it]
Running chain 3: 50%|████████████████████████████▌ | 1500/3000
[2:16:16<2:26:33, 5.86s/it]
Running chain 0: 60%|██████████████████████████████████▏ | 1800/3000
[2:24:33<1:51:53, 5.59s/it]
Running chain 2: 60%|██████████████████████████████████▏ | 1800/3000
[2:24:37<1:41:07, 5.06s/it]
Running chain 3: 55%|███████████████████████████████▎ | 1650/3000
[2:25:46<1:57:43, 5.23s/it]
Running chain 0: 65%|█████████████████████████████████████ | 1950/3000
[2:35:20<1:31:06, 5.21s/it]
Running chain 0: 70%|███████████████████████████████████████▉ | 2100/3000
[2:42:43<1:07:53, 4.53s/it]
Running chain 2: 70%|███████████████████████████████████████▉ | 2100/3000
[2:44:22<1:08:00, 4.53s/it]
Running chain 0: 75%|████████████████████████████████████████████▎ |
2250/3000 [2:48:35<48:20, 3.87s/it]
Running chain 0: 80%|███████████████████████████████████████████████▏ |
2400/3000 [2:56:46<36:53, 3.69s/it]
Running chain 3: 75%|████████████████████████████████████████████▎ |
2250/3000 [2:57:28<42:04, 3.37s/it]
Running chain 2: 75%|██████████████████████████████████████████▊ |
2250/3000 [2:58:24<1:00:43, 4.86s/it]
Running chain 0: 85%|██████████████████████████████████████████████████▏ |
2550/3000 [3:03:07<25:04, 3.34s/it]
Running chain 0: 90%|█████████████████████████████████████████████████████
| 2700/3000 [3:07:37<14:23, 2.88s/it]
Running chain 3: 90%|█████████████████████████████████████████████████████
| 2700/3000 [3:13:57<13:13, 2.64s/it]
Running chain 0:
95%|████████████████████████████████████████████████████████ | 2850/3000
[3:14:41<07:09, 2.86s/it]
Running chain 0:
100%|███████████████████████████████████████████████████████████| 3000/3000
[3:19:53<00:00, 2.63s/it]
Running chain 3:
100%|███████████████████████████████████████████████████████████| 3000/3000
[3:22:46<00:00, 2.15s/it]
Running chain 2: 85%|██████████████████████████████████████████████████▏ |
2550/3000 [3:23:32<36:06, 4.81s/it]
Running chain 2: 90%|█████████████████████████████████████████████████████
| 2700/3000 [3:28:56<20:05, 4.02s/it]
Running chain 2:
95%|████████████████████████████████████████████████████████ | 2850/3000
[3:34:30<08:41, 3.48s/it]
Running chain 0:
100%|███████████████████████████████████████████████████████████| 3000/3000
[3:39:33<00:00, 4.39s/it]
Running chain 1:
100%|███████████████████████████████████████████████████████████| 3000/3000
[3:39:33<00:00, 4.39s/it]
Running chain 2:
100%|███████████████████████████████████████████████████████████| 3000/3000
[3:39:33<00:00, 4.39s/it]
Running chain 3:
100%|███████████████████████████████████████████████████████████| 3000/3000
[3:39:33<00:00, 4.39s/it]
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68:
UserWarning: Explicitly requested dtype float64 requested in astype is not
available, and will be truncated to dtype float32. To enable more dtypes,
set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell
environment variable. See https://github.com/google/jax#current-gotchas
for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
There were 4000 divergences after tuning. Increase target_accept or
reparameterize.
The rhat statistic is larger than 1.01 for some parameters. This indicates
problems during sampling. See https://arxiv.org/abs/1903.08008 for details
The effective sample size per chain is smaller than 100 for some
parameters. A higher number is needed for reliable rhat and ess
computation. See https://arxiv.org/abs/1903.08008 for details
—
Reply to this email directly, view it on GitHub
<#565>, or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AAG7TFGNLDTWSAFMXXHQROTZSOPRZAVCNFSM6AAAAABM2TXQ22VHI2DSMVQWIX3LMV43ERDJONRXK43TNFXW4OZXGA3TMNBQGU>
.
You are receiving this because you are subscribed to this thread.Message
ID: ***@***.***>
|
Beta Was this translation helpful? Give feedback.
-
|
Thank you so much for the timely response. What do those hyperparameters mean, and how to decide what are the optimized numbers for them? I'll change t to only with random effect and remove the covariates for fix effect. And see if that helps with convergence. |
Beta Was this translation helpful? Give feedback.
-
|
Still can't converge infer_model_0 = model_0.sample( C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more. |
Beta Was this translation helpful? Give feedback.
-
for a model like this:
model_0 = hssm.HSSM(data=data,
hierarchical=True,
prior_settings="safe",
include=[
{
"name": "v",
"formula": "v ~ condition + diff + (1|participant_id)",
},
{
"name": "a",
"formula": "a ~ condition + diff + (1|participant_id)",
},
{
"name": "t",
"formula": "t ~ condition + diff + (1|participant_id)",
},
],
)
Are these hyperparameters enough to make it converge?
infer_model_0 = model_0.sample(
sampler="nuts_numpyro", chains=4, cores=1, draws=1000, tune=2000, target_accept=0.95
)
Sometimes I get error messages like this, that says target_accept has to be higher, but how high is high enough, how many chains, cares, draws, tune are enough for how big of a dataset? Any advice on how to set it up?
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
Compiling.. : 0%| | 0/3000 [00:00<?, ?it/s]
0%| | 0/3000 [00:00<?, ?it/s]
Compiling.. : 0%| | 0/3000 [00:00<?, ?it/s]
0%| | 0/3000 [00:00<?, ?it/s]
Compiling.. : 0%| | 0/3000 [00:00<?, ?it/s]
0%| | 0/3000 [00:00<?, ?it/s]
Compiling.. : 0%| | 0/3000 [00:00<?, ?it/s]C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
Running chain 3: 0%| | 0/3000 [00:09<?, ?it/s]
Running chain 0: 0%| | 0/3000 [00:09<?, ?it/s]
Running chain 1: 0%| | 0/3000 [00:09<?, ?it/s]
Running chain 3: 5%|███ | 150/3000 [07:45<2:24:14, 3.04s/it]
Running chain 1: 5%|███ | 150/3000 [08:27<2:37:34, 3.32s/it]
Running chain 0: 5%|███ | 150/3000 [10:53<3:23:53, 4.29s/it]
Running chain 3: 10%|██████ | 300/3000 [13:51<2:00:54, 2.69s/it]
Running chain 0: 10%|██████ | 300/3000 [17:53<2:33:35, 3.41s/it]
Running chain 1: 15%|█████████ | 450/3000 [21:10<1:53:13, 2.66s/it]
Running chain 0: 15%|█████████ | 450/3000 [25:47<2:20:09, 3.30s/it]
Running chain 0: 20%|████████████ | 600/3000 [32:32<2:02:25, 3.06s/it]
Running chain 2: 15%|█████████ | 450/3000 [34:05<3:10:18, 4.48s/it]
Running chain 1: 20%|████████████ | 600/3000 [36:07<2:38:51, 3.97s/it]
Running chain 0: 25%|███████████████ | 750/3000 [41:29<2:01:51, 3.25s/it]
Running chain 2: 20%|████████████ | 600/3000 [43:04<2:45:10, 4.13s/it]
Running chain 3: 25%|███████████████ | 750/3000 [44:13<2:03:22, 3.29s/it]
Running chain 1: 25%|███████████████ | 750/3000 [46:10<2:29:38, 3.99s/it]
Running chain 1: 30%|██████████████████ | 900/3000 [53:58<2:09:18, 3.69s/it]
Running chain 2: 25%|███████████████ | 750/3000 [57:48<2:58:38, 4.76s/it]
Running chain 0: 30%|█████████████████▍ | 900/3000 [1:02:30<2:55:04, 5.00s/it]
Running chain 2: 30%|█████████████████▍ | 900/3000 [1:06:52<2:33:12, 4.38s/it]
Running chain 0: 35%|███████████████████▉ | 1050/3000 [1:11:39<2:28:19, 4.56s/it]
Running chain 1: 45%|█████████████████████████▋ | 1350/3000 [1:18:01<1:33:06, 3.39s/it]
Running chain 3: 30%|█████████████████▍ | 900/3000 [1:24:22<4:27:07, 7.63s/it]
Running chain 1: 50%|████████████████████████████▌ | 1500/3000 [1:25:59<1:23:06, 3.32s/it]
Running chain 0: 40%|██████████████████████▊ | 1200/3000 [1:29:24<2:41:09, 5.37s/it]
Running chain 1: 55%|███████████████████████████████▎ | 1650/3000 [1:32:28<1:09:48, 3.10s/it]
Running chain 3: 35%|███████████████████▉ | 1050/3000 [1:35:09<3:32:44, 6.55s/it]
Running chain 0: 45%|█████████████████████████▋ | 1350/3000 [1:39:59<2:17:56, 5.02s/it]
Running chain 1: 60%|██████████████████████████████████▏ | 1800/3000 [1:44:49<1:13:11, 3.66s/it]
Running chain 0: 50%|████████████████████████████▌ | 1500/3000 [1:48:15<1:52:12, 4.49s/it]
Running chain 3: 40%|██████████████████████▊ | 1200/3000 [1:49:58<3:10:28, 6.35s/it]
Running chain 1: 65%|█████████████████████████████████████ | 1950/3000 [1:53:38<1:03:21, 3.62s/it]
Running chain 2: 50%|████████████████████████████▌ | 1500/3000 [1:56:49<1:51:53, 4.48s/it]
Running chain 1: 70%|█████████████████████████████████████████▎ | 2100/3000 [1:57:13<44:23, 2.96s/it]
Running chain 3: 45%|█████████████████████████▋ | 1350/3000 [2:00:49<2:37:19, 5.72s/it]
Running chain 1: 75%|████████████████████████████████████████████▎ | 2250/3000 [2:02:02<33:06, 2.65s/it]
Running chain 1: 80%|███████████████████████████████████████████████▏ | 2400/3000 [2:05:21<22:30, 2.25s/it]
Running chain 1: 85%|██████████████████████████████████████████████████▏ | 2550/3000 [2:08:28<14:37, 1.95s/it]
Running chain 2: 55%|███████████████████████████████▎ | 1650/3000 [2:10:12<1:46:44, 4.74s/it]
Running chain 1: 90%|█████████████████████████████████████████████████████ | 2700/3000 [2:11:23<08:34, 1.71s/it]
Running chain 0: 55%|███████████████████████████████▎ | 1650/3000 [2:15:23<2:24:47, 6.44s/it]
Running chain 3: 50%|████████████████████████████▌ | 1500/3000 [2:16:16<2:26:33, 5.86s/it]
Running chain 0: 60%|██████████████████████████████████▏ | 1800/3000 [2:24:33<1:51:53, 5.59s/it]
Running chain 2: 60%|██████████████████████████████████▏ | 1800/3000 [2:24:37<1:41:07, 5.06s/it]
Running chain 3: 55%|███████████████████████████████▎ | 1650/3000 [2:25:46<1:57:43, 5.23s/it]
Running chain 0: 65%|█████████████████████████████████████ | 1950/3000 [2:35:20<1:31:06, 5.21s/it]
Running chain 0: 70%|███████████████████████████████████████▉ | 2100/3000 [2:42:43<1:07:53, 4.53s/it]
Running chain 2: 70%|███████████████████████████████████████▉ | 2100/3000 [2:44:22<1:08:00, 4.53s/it]
Running chain 0: 75%|████████████████████████████████████████████▎ | 2250/3000 [2:48:35<48:20, 3.87s/it]
Running chain 0: 80%|███████████████████████████████████████████████▏ | 2400/3000 [2:56:46<36:53, 3.69s/it]
Running chain 3: 75%|████████████████████████████████████████████▎ | 2250/3000 [2:57:28<42:04, 3.37s/it]
Running chain 2: 75%|██████████████████████████████████████████▊ | 2250/3000 [2:58:24<1:00:43, 4.86s/it]
Running chain 0: 85%|██████████████████████████████████████████████████▏ | 2550/3000 [3:03:07<25:04, 3.34s/it]
Running chain 0: 90%|█████████████████████████████████████████████████████ | 2700/3000 [3:07:37<14:23, 2.88s/it]
Running chain 3: 90%|█████████████████████████████████████████████████████ | 2700/3000 [3:13:57<13:13, 2.64s/it]
Running chain 0: 95%|████████████████████████████████████████████████████████ | 2850/3000 [3:14:41<07:09, 2.86s/it]
Running chain 0: 100%|███████████████████████████████████████████████████████████| 3000/3000 [3:19:53<00:00, 2.63s/it]
Running chain 3: 100%|███████████████████████████████████████████████████████████| 3000/3000 [3:22:46<00:00, 2.15s/it]
Running chain 2: 85%|██████████████████████████████████████████████████▏ | 2550/3000 [3:23:32<36:06, 4.81s/it]
Running chain 2: 90%|█████████████████████████████████████████████████████ | 2700/3000 [3:28:56<20:05, 4.02s/it]
Running chain 2: 95%|████████████████████████████████████████████████████████ | 2850/3000 [3:34:30<08:41, 3.48s/it]
Running chain 0: 100%|███████████████████████████████████████████████████████████| 3000/3000 [3:39:33<00:00, 4.39s/it]
Running chain 1: 100%|███████████████████████████████████████████████████████████| 3000/3000 [3:39:33<00:00, 4.39s/it]
Running chain 2: 100%|███████████████████████████████████████████████████████████| 3000/3000 [3:39:33<00:00, 4.39s/it]
Running chain 3: 100%|███████████████████████████████████████████████████████████| 3000/3000 [3:39:33<00:00, 4.39s/it]
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
C:\Users\jc4472\AppData\Local\anaconda3\envs\hssm_env\Lib\site-packages\jax_src\numpy\array_methods.py:68: UserWarning: Explicitly requested dtype float64 requested in astype is not available, and will be truncated to dtype float32. To enable more dtypes, set the jax_enable_x64 configuration option or the JAX_ENABLE_X64 shell environment variable. See https://github.com/google/jax#current-gotchas for more.
return lax_numpy.astype(arr, dtype, copy=copy, device=device)
There were 4000 divergences after tuning. Increase
target_acceptor reparameterize.The rhat statistic is larger than 1.01 for some parameters. This indicates problems during sampling. See https://arxiv.org/abs/1903.08008 for details
The effective sample size per chain is smaller than 100 for some parameters. A higher number is needed for reliable rhat and ess computation. See https://arxiv.org/abs/1903.08008 for details
Beta Was this translation helpful? Give feedback.
All reactions