-
Hi,
I am not personally having an issue, but I wanted to make you aware of some things in the tutorial that might throw off a new user.
When the tutorial starts talking about the high level f…
-
This is meant to be a sketch of an idea how we could visualize the evaluation of an execution tree. Let's assume we have a simple expression `store(A, add(B, C))`, then the following visualization all…
-
expected (new) date : 2017-04-01
expected features:
- Pandas-0.22.0
- PyQt5-5.9.1-5.9.2, Spyder-3.2.8
- Bokeh-0.12.15, Holoviews-1.10.0dev1 (jupyterlab compatibility)
- Notebook-5.4.1 (Term…
-
Hi,
in the appendix of [Attention Is All You Need](https://arxiv.org/pdf/1706.03762.pdf) some very nice visualizations of the attention mechanism can be seen.
How can I create those visualizatio…
-
I'm planning on writing a lightweight extension to PyTorch. The goal is to speed up my research in 2 ways
1) Testing many models should be easy, and it should be easy to keep track of the parameters …
-
This is a feature request. I would like to find out if my model fit is converging on a hold out set. Currently eval_every operates on the latest mini-batch. A parameter ( perhaps called eval_corpus ) …
-
I have followed the instructions and changed the example on two computers:
# sys.path.append('/home/evjang/thesis/tensor_debugger')
# sys.path.append('/home/lee/softwareinstalled/anaconda3-5/tdb_ext')…
-
in the keras/keras/callbacks.py class TensorBoard
` embeddings = {layer.name: layer.weights[0]
for layer in self.model.layers
if la…
-
release estimate : end-october 2017
features:
- WinPython-3.6 is the default download
- matplotlib-2.1.0, spyder-3.2.4, scikit-learn-0.19.1, pandas-0.21.0
- updated notebook experience: Ipython-…
-
Has there been discussion of exporting tensorboard visualizations as images (e.g. `.png`)? This would be helpful for rough drafts and work-in-progress type presentations.