-
**Is your feature request related to a problem? Please describe.**
The multigpu criteo benchmark is hardcoding the best number of hash partitions for each categorical variable:
https://github.com/…
-
### Checklist
- [X] I agree to the terms within the [OpenFGA Code of Conduct](https://github.com/openfga/.github/blob/main/CODE_OF_CONDUCT.md).
### Describe the problem you'd like to have solved
I'…
-
https://github.com/criteo-research/reco-gym/blob/f8553d197f42ec2f415aefce48525d0e9b10ddaa/recogym/envs/reco_env_v1.py#L55-L57
Should the line 56th of the file "reco_env_v1.py" be
`[self.config.pr…
-
### 🔖 Enhancement description
Look at the amount of CVEs from **ImageMagick**:
https://cve.mitre.org/cgi-bin/cvekey.cgi?keyword=Imagemagick
Total of **672** CVEs
Compared to **libvips**:
https://…
-
We could either use a pre-existing tool or do our own implementation.
**Pre-existing tools:** (Note, openapi and JSONSchema are different formats that are related):
* https://github.com/Azure/op…
-
I was wondering if this plugin was already updated for marathon 1.11. It looks like this 1.6.352 is not working anymore. I think 1.7.202 is in the maven repository, do not know if newer ones are there…
-
get_criteo_feature.py中的特征下标应该是从1开始的,PNN.py中直接使用idx,这样的话使用tf.nn.embedding_lookup进行embedding是从0开始?
-
I am finding issues with preprocessing Criteo Terabyte Dataset for the inference runs for Intel CPU.
When the readme file talks about to prepare the DLRM dataset,
```
Create a directory (such a…
-
-
Hi, Thanks for sharing your code, really helpful.
I have been able to reproduce results on custom subset of Frappe and Avazu datasets, however on Criteo dataset, am getting NaN/None on numerics_check…