lanlin

Some issues on remote interpreter

3

## Expected Behaviour The remote interpreter works normally in the latest anaconda. ## Actual Behaviour As reported in https://github.com/DamnWidget/anaconda/issues/872#issuecomment-714366984 @scuml @DamnWidget After some trials, I found that the remote interpreter...

How can we define a list of integers?

1

In many cases, we need pass a list integers to a program, such as a list containing the number of units per layer in a neural network. For now, with...

version check makes "from ogb.xxx import xxx" get stuck

2

@weihua916 @rusty1s The version check in [version.py](https://github.com/snap-stanford/ogb/blob/master/ogb/version.py) makes `from ogb.xxx import xxx` get stuck. With those code: ``` ❯ ipython Python 3.9.12 (main, Apr 5 2022, 06:56:58) Type 'copyright', 'credits'...

I found that we need to [explicitly pass](https://github.com/deepmind/dm-haiku/blob/2a6c03406fd1281bae104b737601faaa37b656e6/haiku/_src/nets/mlp.py#L103-L108) the argument `rng` to the `__call__` function of MLP when dropout is used. If we have a nested function that contains an...

Hi everyone, I'm trying to re-implement the [GMN model](https://github.com/deepmind/deepmind-research/tree/master/graph_matching_networks) from deepmind in Jax. This model is designed to compute the similarity score between two graphs, and needs to compute the...

Hi @xiangyue9607 Thanks for releasing these datasets. Recently, I try to use Mashup_PPI to do some experiments, and find that there are many duplicate edges in the edgelist. Perhaps it...

Doubt about featureless dataset

5

Hi Thomas, Thanks for sharing your code. I want to use GCN for datasets without features. From previous issues, you mentioned that one option is to use embeddings learned by...

About loss design

1

Hi @tkipf . I saw a comment that in practice we can use the representations learned at each layer to construct a compound loss function, e.g., the summation of three...

[ogb_lsc] the code for pcq does not use the train split during training

1

In the code for the dataset PCQM4M, the function [`_load_smiles`](https://github.com/deepmind/deepmind-research/blob/6fcb84268e74af981ae1496bfc2cb9ba9d701ef2/ogb_lsc/pcq/dataset_utils.py#L279-L298) is used to load data when building data iterator. However, when the `split` is set to `train`, the function [`datasets.load_all_except_kth_fold_indices`](https://github.com/deepmind/deepmind-research/blob/6fcb84268e74af981ae1496bfc2cb9ba9d701ef2/ogb_lsc/pcq/dataset_utils.py#L290)...