Juncheng Liu

Parameter Efficiency of ConvE

3

In your paper, you claim ConvE uses less parameter compared with DistMult. But I think in your code DistMult only uses O(num_entities*emdding_dim + num_rels*embedding_dim) and ConvE uses more parameters. I...

I am confused about the noise you used in the code. In your writeup at arxiv, you said you used parameter space noise(https://arxiv.org/abs/1706.01905). However, I think you actually used NoisyNet...

Instructions for running the model

6

Could you provide more detailed instructions for running the model? Thank you!

Why the baseline performance is different with the original version?

1

In the original version of [GraphSAGE](https://github.com/williamleif/GraphSAGE/tree/master/graphsage), the micro f1 score after 50 epochs in PPI dataset is only around 0.57. I don't modify the args in the shell script. In...

update from upstream & make the implement more robust and meaningful in DP/Policy Evaluation Solution

1

1. update the description. 2. make it more robust. I think in the for loop 'for prob, next_state, reward, done in env.P[s][a]:', we need to firstly sum the values from...

Created a new callback to save the checkpoints as in huggingface format. Basically, we inherited [ModelCheckpoint](https://github.com/Lightning-AI/pytorch-lightning/blob/025c30e6bc9c4e5cda0569b85db6993842413605/src/lightning/pytorch/callbacks/model_checkpoint.py#L50) from pytorch lightning and override the inherent functions `_save_checkpoint` and `_remove_checkpoint`.

Loading checkpoints from local files

7

**Is your feature request related to a problem? Please describe.** We faced a problem when try to load the checkpoints from local files. Say using a model yaml under `eval`...

**Describe the bug** With the current `datasets` version, it cannot download cloudops datasets. **To Reproduce** ```python from datasets import load_dataset dataset = load_dataset('Salesforce/cloudops_tsf', 'azure_vm_traces_2017') ``` **Expected behavior** It should successfully...