friendliai/periflow-python-sdk issues and pull requests

#57 - Add a deprecation notice to README.

Pull Request - State: open - Opened by kooyunmo over 1 year ago

#56 - [NVIDIA] Add FP8 tutorial using TransformerEngine

Pull Request - State: closed - Opened by ksivaman over 1 year ago - 3 comments
Labels: enhancement

#55 - Update VERSION

Pull Request - State: closed - Opened by kooyunmo almost 2 years ago

#54 - [PF-1189] allow checkpoint uploading only once for an iteration

Pull Request - State: closed - Opened by ktaebum almost 2 years ago

#53 - Megatron example

Pull Request - State: closed - Opened by ktaebum about 2 years ago

#52 - README for all examples

Pull Request - State: open - Opened by footprinthere over 2 years ago

#51 - MMF VLM example

Pull Request - State: open - Opened by footprinthere over 2 years ago - 1 comment

#50 - Huggingface speech recognition finetuning examples

Pull Request - State: closed - Opened by footprinthere over 2 years ago - 3 comments

#49 - Huggingface LoRA finetuning example

Pull Request - State: closed - Opened by footprinthere over 2 years ago - 1 comment

#48 - [PF-62] Handle torch.multiprocessing.spawn use case

Pull Request - State: closed - Opened by ktaebum over 2 years ago

#47 - detectron2/mmdetection object detection

Pull Request - State: closed - Opened by footprinthere over 2 years ago - 4 comments

#46 - TIMM/Torchvision ImageNet

Pull Request - State: closed - Opened by footprinthere over 2 years ago

#45 - HF Transformers seq-to-seq summarization finetuning

Pull Request - State: closed - Opened by footprinthere over 2 years ago

#44 - Lightning Transformers MLM Pre-training

Pull Request - State: closed - Opened by footprinthere over 2 years ago

#43 - Lightning Transformers Auto-regressive LM Pre-training

Pull Request - State: closed - Opened by footprinthere over 2 years ago - 3 comments

#42 - Example

Pull Request - State: closed - Opened by footprinthere over 2 years ago - 1 comment

#41 - Example

Pull Request - State: closed - Opened by footprinthere over 2 years ago

#40 - version up

Pull Request - State: closed - Opened by ktaebum over 2 years ago

#39 - add trigger_time of checkpoint_upload

Pull Request - State: closed - Opened by ktaebum over 2 years ago

#38 - Last step fix

Pull Request - State: closed - Opened by ktaebum over 2 years ago

#37 - Add copyright and change emergency save condition

Pull Request - State: closed - Opened by kooyunmo over 2 years ago

#36 - Use `NPROC_PER_NODE` auto-set env var in examples

Pull Request - State: closed - Opened by kooyunmo over 2 years ago

#35 - vm type fix

Pull Request - State: closed - Opened by ktaebum over 2 years ago

#34 - Update VERSION

Pull Request - State: closed - Opened by Algy over 2 years ago

#33 - Hotfix 0.1.0-rc1

Pull Request - State: closed - Opened by Algy over 2 years ago

#32 - add readme

Pull Request - State: closed - Opened by ktaebum over 2 years ago - 1 comment

#31 - Freeze repository for package version 0.1.0-rc1

Pull Request - State: closed - Opened by Algy over 2 years ago

#30 - Examples

Pull Request - State: closed - Opened by ktaebum over 2 years ago

#29 - remove job_finished

Pull Request - State: closed - Opened by ktaebum over 2 years ago

#28 - fix getting RANK

Pull Request - State: closed - Opened by ktaebum over 2 years ago

#27 - cleanup sdk

Pull Request - State: closed - Opened by ktaebum over 2 years ago

#26 - reorder file directory

Pull Request - State: closed - Opened by ktaebum over 2 years ago

#25 - Add configuration dotfiles

Pull Request - State: closed - Opened by kooyunmo over 2 years ago

#24 - refactoring sdk

Pull Request - State: closed - Opened by ktaebum over 2 years ago - 1 comment

#23 - not always create cloud_path

Pull Request - State: closed - Opened by ktaebum over 2 years ago

#22 - do not send save_type when not saved

Pull Request - State: closed - Opened by ktaebum over 2 years ago

#21 - [PFT-304] update ipc

Pull Request - State: closed - Opened by ktaebum over 2 years ago - 1 comment

#20 - Send last_step to WorkerFTModule

Pull Request - State: closed - Opened by ktaebum over 2 years ago

#19 - make run when CKPT_DIR is not set

Pull Request - State: closed - Opened by ktaebum almost 3 years ago

#18 - [PFT-271] Fix pf.load() so that it supports all the arguments of torch.load()

Pull Request - State: closed - Opened by DifferentSC almost 3 years ago - 1 comment

#17 - [PFT-272] Calculate parallelism rank in sdk

Pull Request - State: closed - Opened by ktaebum almost 3 years ago

#16 - [PFT-265] Add rank information in SDK metrics

Pull Request - State: closed - Opened by DifferentSC almost 3 years ago

#15 - [PFT-256] Make tests for failure paths of SDK

Pull Request - State: closed - Opened by DifferentSC almost 3 years ago

#14 - [PFT-254] Fix save error in SDK

Pull Request - State: closed - Opened by DifferentSC almost 3 years ago

#13 - [PFT-222, PFT-224] Newly designed Periflow Python SDK

Pull Request - State: closed - Opened by DifferentSC almost 3 years ago

#12 - [PFT-175] Minor changes in SDK

Pull Request - State: closed - Opened by ktaebum almost 3 years ago

#11 - Fix DDP download problem

Pull Request - State: closed - Opened by DifferentSC about 3 years ago - 3 comments

#10 - Update transformers example & Add wandb script

Pull Request - State: closed - Opened by kooyunmo about 3 years ago

#9 - Huggingface Transformers example

Pull Request - State: closed - Opened by Hyeonmin-Ha about 3 years ago

#8 - Add docker script

Pull Request - State: closed - Opened by sj6077 about 3 years ago

#7 - Add unit tests for basic operations

Issue - State: closed - Opened by DifferentSC about 3 years ago

#6 - Automatic recovery and resumable iterator

Pull Request - State: closed - Opened by DifferentSC about 3 years ago - 2 comments

#5 - Automatic Recovery from Checkpoints

Issue - State: closed - Opened by DifferentSC about 3 years ago

#4 - Support PyTorch DDP

Issue - State: closed - Opened by DifferentSC about 3 years ago

#3 - Resumable Iterator for Data Loading

Issue - State: closed - Opened by DifferentSC about 3 years ago

#2 - Initial Interface & Implmentation for Periflow

Issue - State: closed - Opened by DifferentSC about 3 years ago - 1 comment

#1 - Initial Interface & Implementation for Periflow Python SDK

Pull Request - State: closed - Opened by DifferentSC about 3 years ago - 2 comments

GitHub / friendliai/periflow-python-sdk issues and pull requests