-
Hello,
I am struggling to trace BLIP2 model from transformers library using `torch_neuronx` to make it work on an inf2. The model I want to trace is the [XXL version](https://huggingface.co/Salesfo…
-
Hi, I'm just hoping to get some attention to this question, if there are any answers, if this is expected behavior, or if this indicates some underlying issue that could be fixed. Original post below …
-
Hi all,
I want to use EBM as a GAM to replace the fully connected layer at the end of a large CNN/Transformer to get interpretable output. However, I need to train the EBM like a deep learning model,…
-
Hi there,
I noticed another critical bug (at least from mine point of view): after LoRa training and even with `do_sample` is False, consecutive inference results in different results:
Loading t…
-
Is it the following behaviour expected? I start tpot with a fixed template "Selector-Transformer-Regressor" but then I get a StackingEstimator model. I am not completely sure, but I think I never expe…
-
Hello,
There are a number of models I'd like to try which require this. I know that I asked you about this in the past, and IIRC you mentioned that you removed it because you wanted to implement i…
-
(From https://github.com/scikit-learn/scikit-learn/pull/27800#issuecomment-1878709518 by @ogrisel)
The proposal/idea is to allow `y` to not be on the same device (and namespace?) as `X` when using …
-
I ran these commands in Google Colab with GPU
```
!wget http://boston.lti.cs.cmu.edu/luyug/coil/msmarco-psg/psg-train.tar.gz
!tar xfz psg-train.tar.gz
!git clone https://github.com/luyug/COIL
!…
-
### Feature request
Export to onnx fails for opset 9 with T5
### Motivation
ONNX opset 9 is required by SNPE, Qualcomm SDK accelerator. By supporting ONNX opset 9, we will unleash ML on the e…
-
Using a well-crafted FAUXPILOT, we can execute inference tasks based on the Codegen model. I read recently that I can work on Fine-tune using the Codegen model on the following website.
* https://new…