mila-iqia/milabench issues and pull requests

#309 - H100

Pull Request - State: open - Opened by Delaunay 27 days ago

#308 - Multirun system

Pull Request - State: open - Opened by Delaunay about 1 month ago

#307 - [HPU] Jax

Issue - State: open - Opened by Delaunay about 1 month ago

#306 - [HPU] torch geometric

Issue - State: open - Opened by Delaunay about 1 month ago
Labels: HPU, custom kernels

#305 - [HPU] llm-lora-mp-gpus

Issue - State: closed - Opened by Delaunay about 1 month ago
Labels: HPU, torchtune

#304 - [rocm] rlhf

Issue - State: open - Opened by Delaunay about 1 month ago
Labels: rocm

#303 - [HPU] rlhf

Issue - State: open - Opened by Delaunay about 2 months ago - 4 comments
Labels: HPU

#302 - lightning

Issue - State: closed - Opened by Delaunay about 2 months ago
Labels: HPU

#301 - [HPU] reformer

Issue - State: closed - Opened by Delaunay about 2 months ago - 1 comment
Labels: HPU

#300 - [HPU] llava-single

Issue - State: closed - Opened by Delaunay about 2 months ago - 2 comments
Labels: HPU

#299 - [HPU] llm-full-mp-gpus

Issue - State: open - Opened by Delaunay about 2 months ago - 2 comments
Labels: HPU, torchtune

#298 - [HPU] llm-lora-ddp-gpus

Issue - State: closed - Opened by Delaunay about 2 months ago - 1 comment
Labels: HPU, torchtune

#297 - llm-lora-single on HPU

Issue - State: closed - Opened by Delaunay about 2 months ago - 1 comment
Labels: HPU

#296 - Torch Geometric on ROCm

Issue - State: closed - Opened by Delaunay about 2 months ago
Labels: rocm

#295 - Jax on ROCm GPUs

Issue - State: closed - Opened by Delaunay about 2 months ago - 1 comment
Labels: rocm

#294 - vjepa-single HPU

Issue - State: closed - Opened by Delaunay about 2 months ago - 5 comments
Labels: HPU

#293 - Rocm

Pull Request - State: open - Opened by Delaunay about 2 months ago

#292 - Hpu

Pull Request - State: open - Opened by Delaunay about 2 months ago

#291 - Staging

Pull Request - State: closed - Opened by Delaunay 2 months ago - 1 comment

#290 - Fix rlhf on trl v0.11.0

Pull Request - State: closed - Opened by satyaog 2 months ago

#289 - Fix llm with torchtune v0.3

Pull Request - State: closed - Opened by satyaog 2 months ago - 9 comments

#288 - Force exaclty one monitor tag

Pull Request - State: closed - Opened by satyaog 2 months ago

#287 - Fix vjepa reporting an error on closing

Pull Request - State: closed - Opened by satyaog 2 months ago

#286 - Batch resizing

Pull Request - State: closed - Opened by Delaunay 2 months ago

#285 - Merged

Pull Request - State: closed - Opened by Delaunay 2 months ago

#284 - Geo gnn fixes

Pull Request - State: closed - Opened by bouthilx 2 months ago

#283 - Staging

Pull Request - State: closed - Opened by Delaunay 3 months ago - 8 comments

#282 - rank 0 is not on the local node

Issue - State: closed - Opened by Delaunay 3 months ago - 2 comments

#281 - diffusion-nodes can fail loading the checkpoint on some rank

Issue - State: open - Opened by Delaunay 3 months ago - 6 comments

#280 - test-ci 2n:2g

Pull Request - State: open - Opened by satyaog 3 months ago

#279 - Fixes loss NaN issue for LlaVa

Pull Request - State: closed - Opened by rabiulcste 3 months ago

#278 - Fix llama3 generation

Pull Request - State: closed - Opened by satyaog 3 months ago

#277 - Vjepa

Pull Request - State: closed - Opened by Delaunay 3 months ago - 2 comments

#276 - Revert "new RLHF benchmark (#273)"

Pull Request - State: closed - Opened by Delaunay 3 months ago - 1 comment

#275 - Rlhf hf

Pull Request - State: closed - Opened by Delaunay 3 months ago

#274 - Added jepa as new benchmark, but require further debug

Pull Request - State: closed - Opened by lezhang7 3 months ago - 1 comment

#273 - new RLHF benchmark

Pull Request - State: closed - Opened by Delaunay 3 months ago - 1 comment

#272 - Update pins

Pull Request - State: closed - Opened by Delaunay 3 months ago

#271 - test-ci 2n:4g

Pull Request - State: open - Opened by satyaog 3 months ago

#270 - test-ci 1n:4g

Pull Request - State: open - Opened by satyaog 3 months ago

#269 - Staging

Pull Request - State: closed - Opened by Delaunay 3 months ago - 7 comments

#268 - Attempt fix on dinov2-giant-nodes

Pull Request - State: closed - Opened by satyaog 3 months ago - 1 comment

#267 - Fix diffusion

Pull Request - State: closed - Opened by satyaog 3 months ago

#266 - Adding LlaVa

Pull Request - State: closed - Opened by rabiulcste 3 months ago - 2 comments

#265 - Staging

Pull Request - State: closed - Opened by Delaunay 3 months ago - 1 comment

#264 - Rl argparse

Pull Request - State: closed - Opened by Delaunay 3 months ago

#263 - Cleanrl jax

Pull Request - State: closed - Opened by Delaunay 3 months ago

#262 - RLHF torchtune

Pull Request - State: open - Opened by Delaunay 3 months ago

#261 - Tweaks 3

Pull Request - State: closed - Opened by Delaunay 3 months ago

#260 - Tweaks 2

Pull Request - State: closed - Opened by Delaunay 3 months ago - 2 comments

#259 - Update pins for CUDA

Pull Request - State: closed - Opened by Delaunay 3 months ago - 1 comment

#258 - Jax & CUDA

Issue - State: closed - Opened by Delaunay 3 months ago

#257 - test-ci

Pull Request - State: open - Opened by satyaog 3 months ago - 2 comments

#256 - [rocm] amdsmi not found

Issue - State: open - Opened by Delaunay 3 months ago
Labels: rocm

#255 - Use llama3 for inference

Issue - State: open - Opened by Delaunay 3 months ago

#254 - Milabench Size

Issue - State: open - Opened by Delaunay 3 months ago

#253 - Rlhf 2

Pull Request - State: closed - Opened by Delaunay 3 months ago - 2 comments

#252 - Benchmark Batch

Pull Request - State: closed - Opened by Delaunay 3 months ago - 2 comments

#251 - Add Dimenet

Pull Request - State: closed - Opened by Delaunay 3 months ago

#250 - Generate llama instead of downloading it

Pull Request - State: closed - Opened by satyaog 4 months ago - 6 comments

#249 - recursiongfn benchmark

Pull Request - State: closed - Opened by josephdviviano 4 months ago

#248 - Multi node tweaks

Pull Request - State: closed - Opened by Delaunay 4 months ago - 1 comment

#247 - Create Dockerfile-xpu & Dockerfile-hpu

Pull Request - State: open - Opened by Delaunay 4 months ago - 1 comment

#246 - [WIP] Geometric GNN benchmarks

Pull Request - State: closed - Opened by bouthilx 4 months ago - 1 comment

#245 - Initial commit with purejaxrl

Pull Request - State: closed - Opened by dapatil211 4 months ago

#244 - light docker image

Issue - State: open - Opened by Delaunay 4 months ago

#243 - Initial commit Torch_PPO_Cleanrl_Atari_Envpool

Pull Request - State: closed - Opened by roger-creus 4 months ago - 2 comments

#242 - Update recipes.rst

Pull Request - State: closed - Opened by Delaunay 4 months ago

#241 - Create execution_modes.rst

Pull Request - State: closed - Opened by Delaunay 4 months ago - 1 comment

#240 - Llama 3

Pull Request - State: closed - Opened by Delaunay 4 months ago - 2 comments

#239 - Data pack redistribution

Issue - State: open - Opened by Delaunay 4 months ago

#238 - Dino

Pull Request - State: closed - Opened by Delaunay 4 months ago - 1 comment

#237 - WIP: Adding a PureJaxRL example

Pull Request - State: open - Opened by lebrice 4 months ago - 1 comment