This message was deleted.
# ask-for-help
s
This message was deleted.
👀 1
🍱 1
b
Hello @Amar Ramesh Kamat Can you share what type of model with pytorch? Can you share your saving model code?
a
@Bo it’s the same as on the site. I was just trying the example
b
I want to make sure, is this the example you are talking about? https://docs.bentoml.org/en/latest/frameworks/pytorch.html
a
Oh. My bad. I thought I shared the link. It’s https://docs.bentoml.org/en/latest/frameworks/pytorch.html
b
Can you try out the repo example? Meanwhile, I will look into the doc one, if that's incorrect, I will update it
a
Sure. Will do
@Bo Tried the example from git. I see this
Copy code
2022-10-06T08:40:30.081-07:00	return F.linear(input, self.weight, self.bias)

2022-10-06T08:40:30.081-07:00	RuntimeError: mat1 and mat2 shapes cannot be multiplied (10x676 and 6760x50)

2022-10-06T08:40:30.081-07:00	2022-10-06T15:40:30+0000 [ERROR] [api_server:3] Exception on /predict_image [POST] (trace=...,sampled=0)

2022-10-06T08:40:30.081-07:00	Traceback (most recent call last):
I am using the same image
0.png
for testing and using the image input.
I see this error log line a lot. Not sure if its benign
Copy code
[ERROR] [runner:bentomlpytorch:1] The "bentoml.pytorch.load_runner" method is being deprecated. "load_runner" arguments will be ignored. Use `bentoml.pytorch.get("bentomlpytorch").to_runner()` instead
2022-10-06T15:00:21+0000 [ERROR] [runner:bentomlpytorch:1] The "bentoml.pytorch.load_runner" method is being deprecated. "load_runner" arguments will be ignored. Use `bentoml.pytorch.get("bentomlpytorch").to_runner()` instead