-
- CNN addition problem
- Increase amount of training data by a lot
- Move all the constant functions in CNN training into a .py file (or more than one eventually -- e.g. training dataset cr…
-
### Summary
I wanted to ask about a technical aspect of DeePMD. Is there a way by which I can load DPs as TF models and perform backprop externally, like on a jupyter notebook. I am able to use DPs i…
-
### Willingness to contribute
No. I cannot contribute this feature at this time.
### Proposal Summary
I have found that MLFlow meets virtually every use case for model tracking and monitoring at my…
-
I have used YOLOV7 in my project and it gives me good detction results, my supervisor told me to include static validation of the results, and to benefit from an in-depth analysis of performance varia…
-
This is feedback from when trying to implement gru/lstm on CoreML driven by https://github.com/webmachinelearning/webnn/issues/689.
The biases and weights are stacked together for forward and backwa…
-
The convolutional weight and bias names have historical + '1' added to them which is not required. This is also the case in the q and r2 networks.
Plus, there are lots of commented out code that ca…
-
Hi, thanks for developing such a comprehensive tool for using sequence models.
I was wondering where I could find the set of methods I should include to be able to use your functions on a model th…
-
Thanks for sharing this repo, great work!
I trained BYOL on my data and noticed that the weights and biases for BN layers are not updated on the saved model. I used resnet18 without pretrained weig…
-
See [here](https://transformer-circuits.pub/2023/monosemantic-features#:~:text=Specific%20metrics%20from,is%20too%20low.)
To do:
- [ ] Consider the best way to calculate these
- [ ] Log these to …
-
Right now, weights and biases (represented as Frames) are not exported with DL models.
We need to export them as part of model export.
Comming from H2O stream:
Hi I trained a DL model using e…