Nikolai Makhotkin

Basically, this model is imported from PyTorch, the full net class is below: ```python import torch import torch.nn as nn import torch.nn.functional as F class Net(nn.Module): def __init__(self): super(Net, self).__init__()...

So I suppose these `F.max_pool2d` operations are converted to `py_func`

@tjingrant yes, here is it (uploaded to GDrive): onnx model (generated by pytorch) - https://drive.google.com/file/d/13yJYYgQiiqxP8Khm-PZ5Q6JwxLi2w_4A/view?usp=sharing original pytorch model - https://drive.google.com/file/d/11BJOI5ucsSmM-9aZBYVIBcDvf9ILihnU/view?usp=sharing

@tjingrant thanks! I'll try today (it is morning for me now) and will write the results here.

@tjingrant The fix works! I just tested `onnx-tf` on `fix-pool` branch: I converted my onnx model to tensorflow again and model inference works! Now it is able to successfully recognize...

Just tried to convert pretrained `ResNet` (resnet101) model to onnx, then to tensorflow. As @kartk said, there is still `py_func` presented in graph. Is there a way to get rid...

Thanks for the answer! Yes, it would be nice to have an additional option flag which will control this behavior (either to export precisely or not).

I see silimar on ragas 0.2.15 and python 3.12. Running it inside asyncio event loop by calling ragas.evaluate(...) ``` result = evaluate( ^^^^^^^^^ File "/usr/local/lib/python3.12/dist-packages/ragas/_analytics.py", line 227, in wrapper result...

@richness @jjmachan It is not solved. The issue itself is that when you use `uvloop` for running FastAPI and uvicorn, ragas crashes. The solution above suggests to change event loop...

@jjmachan Thank you for the support, that would be awesome!