Error f"{dims_supplied} must be a permuted list of {dims_all}, unless ... is included" - but example worked before
#799
Answered
by
GStechschulte
danieltomasz
asked this question in
Q&A
-
|
I try to run an example that worked before: import requests
import pandas as pd
from io import StringIO
import bambi as bmb
url = 'https://raw.githubusercontent.com/crnolan/pyrba/main/data.txt' # replace with your url
response = requests.get(url)
data = response.text
# Convert the string to a file-like object
data_io = StringIO(data)
# Read the data into a DataFrame
df = pd.read_table(data_io, delimiter=r"\s+")
print(df.head())
model = bmb.Model("y ~ x + (1|subject) + (x|ROI)", df)
results = model.fit(
tune=4000,
draws=1000,
chains=8,
inference_method="numpyro_nuts",
nuts_kwargs=dict(max_tree_depth=100),
)the daframe has following form I am getting this error I am on Python 3.12 |
Beta Was this translation helpful? Give feedback.
Answered by
GStechschulte
Apr 8, 2024
Replies: 1 comment 8 replies
-
|
I've got a reproduction here
|
Beta Was this translation helpful? Give feedback.
8 replies
-
|
I did clean install, both environment have |
Beta Was this translation helpful? Give feedback.
-
|
The git version actually doesnt compute hierarchical model (I am now computing this model while the |
Beta Was this translation helpful? Give feedback.
-
|
Should this be converted into proper issue (wrong posterior dim??), import requests
import pandas as pd
from io import StringIO
import bambi as bmb
url = "https://raw.githubusercontent.com/crnolan/pyrba/main/data.txt" # replace with your url
response = requests.get(url)
data = response.text
# Convert the string to a file-like object
data_io = StringIO(data)
# Read the data into a DataFrame
df = pd.read_table(data_io, delimiter=r"\s+")
print(df.head())
print(df.nunique())
model = bmb.Model("y ~ (1|subject) + (1|ROI)", df)
results = model.fit(
tune=4000,
draws=1000,
chains=8,
inference_method="numpyro_nuts",
max_tree_depth=3,
)the bambi 0.13 (in python 3.11) gives expected results, but not git version (arviz 0.18 in both env) |
Beta Was this translation helpful? Give feedback.
-
|
Yeah, I am thinking opening an issue would be appropriate. Thanks for opening the discussion and giving us these insights! |
Beta Was this translation helpful? Give feedback.
Answer selected by
danieltomasz
-
|
I had copied my examples here #800, thanks! |
Beta Was this translation helpful? Give feedback.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Yeah, I am thinking opening an issue would be appropriate. Thanks for opening the discussion and giving us these insights!