All issues

Search results

Criteo 1TB dataset is corrupted

Status: Open.

#413 In facebookresearch/dlrm;

· Denker233 opened on Apr 6, 2026

Distributed Training Fails

Status: Open.

#412 In facebookresearch/dlrm;

· shirinebadi opened on Mar 9, 2026

How to convert DistributedModelParallel to quantize_inference_model and use torch.jit.script to save?

Status: Open.

#409 In facebookresearch/dlrm;

· intfish123 opened on Nov 19, 2025

If I train a model using TorchRec, how can I export the model to ONNX?

Status: Open.

#408 In facebookresearch/dlrm;

· ziqunyu opened on Oct 17, 2025

DLRM: Inconsistent Network Responses

Status: Open.

#403 In facebookresearch/dlrm;

· TheProdigyLeague opened on Aug 5, 2025

DRLM model for ranking

Status: Open.

#402 In facebookresearch/dlrm;

· ali-fani-sd opened on Jun 23, 2025

Expected Bandwidth for running DLRM

Status: Open.

#400 In facebookresearch/dlrm;

· noahnisbet opened on Jun 10, 2025

DLRM predictive performance vs older models (e.g., XGBoost)

Status: Open.

#396 In facebookresearch/dlrm;

· RyanZotti opened on May 20, 2025

Consistently Getting ~0.5 AUROC on Full Kaggle Criteo Dataset with DLRM (torchrec)

Status: Open.

#395 In facebookresearch/dlrm;

· ali-fani-sd opened on Apr 8, 2025

data_utils.py fails with Numpy 2.0's new handling of overflow error

Status: Open.

#394 In facebookresearch/dlrm;

· kwmaeng91 opened on Apr 2, 2025

Questions about Commands for DLRM Training

Status: Open.

#392 In facebookresearch/dlrm;

· janghobaik opened on Oct 25, 2024

Questions about the necessity of creating a dictionary in data_utils.py

Status: Open.

#391 In facebookresearch/dlrm;

· Arith2 opened on Oct 16, 2024