User contributions for Lucasn1

23 October 2024

• 20:4320:43, 23 October 2024 diff hist +3,005‎ Apptainer ‎No edit summary

17 September 2024

• 18:0518:05, 17 September 2024 diff hist −1‎ m Huggingface ‎ →‎Downloading the Model and the Dataset
• 18:0418:04, 17 September 2024 diff hist −4‎ m Huggingface ‎ →‎Downloading the Model and the Dataset
• 18:0118:01, 17 September 2024 diff hist +12‎ m Huggingface ‎ →‎Downloading the Model and the Dataset
• 18:0018:00, 17 September 2024 diff hist −12‎ m Huggingface ‎ →‎Downloading the Model and the Dataset
• 17:5517:55, 17 September 2024 diff hist +5‎ m Huggingface ‎ →‎Downloading the Model and the Dataset
• 17:5417:54, 17 September 2024 diff hist +7‎ m Huggingface ‎ →‎Downloading the Model and the Dataset
• 17:5217:52, 17 September 2024 diff hist −11‎ m Huggingface ‎ →‎Downloading the Model and the Dataset
• 17:4617:46, 17 September 2024 diff hist −2‎ m Huggingface ‎ →‎Downloading the Model and the Dataset
• 17:0717:07, 17 September 2024 diff hist 0‎ m Huggingface ‎ →‎Training Large Language Models (LLMs)
• 17:0717:07, 17 September 2024 diff hist −3‎ m Huggingface ‎ →‎Training Large Language Models (LLMs)
• 17:0017:00, 17 September 2024 diff hist +6,432‎ Huggingface ‎No edit summary

11 September 2024

• 14:4614:46, 11 September 2024 diff hist −117‎ m PyTorch ‎No edit summary
• 14:4614:46, 11 September 2024 diff hist −117‎ m PyTorch ‎No edit summary
• 14:4514:45, 11 September 2024 diff hist −117‎ m PyTorch ‎No edit summary
• 14:4414:44, 11 September 2024 diff hist −315‎ m PyTorch ‎No edit summary

18 July 2024

• 15:2115:21, 18 July 2024 diff hist +5‎ m Dask ‎ →‎Single Node current

17 July 2024

• 15:3715:37, 17 July 2024 diff hist +54‎ m PyTorch ‎No edit summary

10 May 2024

• 19:3419:34, 10 May 2024 diff hist +11‎ m Weights & Biases (wandb) ‎No edit summary

12 February 2024

• 21:0821:08, 12 February 2024 diff hist +11‎ m Dask ‎ →‎Single Node

17 January 2024

• 15:4015:40, 17 January 2024 diff hist 0‎ m Huggingface ‎ →‎Multi-GPU & multi-node jobs with Accelerate

18 October 2023

• 15:1315:13, 18 October 2023 diff hist +18‎ m PyTorch ‎No edit summary

3 October 2023

• 18:2218:22, 3 October 2023 diff hist +532‎ TensorFlow ‎No edit summary

22 September 2023

• 18:1118:11, 22 September 2023 diff hist +1‎ m Dask ‎No edit summary

19 September 2023

• 15:1715:17, 19 September 2023 diff hist +13‎ m Dask ‎No edit summary

5 July 2023

• 19:0819:08, 5 July 2023 diff hist −14‎ m PyTorch ‎No edit summary
• 19:0619:06, 5 July 2023 diff hist −7‎ m PyTorch ‎No edit summary
• 19:0319:03, 5 July 2023 diff hist −578‎ m PyTorch ‎No edit summary
• 18:5918:59, 5 July 2023 diff hist 0‎ m PyTorch ‎No edit summary
• 18:5918:59, 5 July 2023 diff hist 0‎ m PyTorch ‎No edit summary
• 18:5818:58, 5 July 2023 diff hist −304‎ m PyTorch ‎No edit summary
• 18:5218:52, 5 July 2023 diff hist −480‎ m PyTorch ‎No edit summary

26 June 2023

• 15:3215:32, 26 June 2023 diff hist +376‎ m PyTorch ‎No edit summary

22 June 2023

• 18:0918:09, 22 June 2023 diff hist +91‎ Weights & Biases (wandb) ‎No edit summary

2 June 2023

• 20:2420:24, 2 June 2023 diff hist +4‎ m Dask ‎No edit summary
• 20:1220:12, 2 June 2023 diff hist −1‎ Dask ‎ →‎Multiple Nodes
• 20:0420:04, 2 June 2023 diff hist +5,379‎ N Dask ‎ Created page with "[https://docs.dask.org/en/stable/ Dask] is a flexible library for parallel computing in Python. It provides parallelized NumPy array and Pandas DataFrame objects, and it enables distributed computing in pure Python with access to the PyData stack. ==Installing our wheel== <!--T:15--> The preferred option is to install it using our provided Python [https://pythonwheels.com/ wheel] as follows: :1. Load a Python module, thus <..."

26 May 2023

• 13:0213:02, 26 May 2023 diff hist −5‎ m Deepspeed ‎ →‎Multi-GPU and multi-node jobs with Deepspeed current
• 13:0113:01, 26 May 2023 diff hist −1‎ m Deepspeed ‎ →‎Multi-GPU and multi-node jobs with Deepspeed

25 May 2023

• 18:4118:41, 25 May 2023 diff hist +160‎ m Deepspeed ‎ →‎Multi-GPU and multi-node jobs with Deepspeed
• 18:3918:39, 25 May 2023 diff hist −9‎ m Deepspeed ‎ →‎Multi-GPU and multi-node jobs with Deepspeed
• 18:1618:16, 25 May 2023 diff hist −53‎ m Deepspeed ‎ →‎Multi-GPU and multi-node jobs with Deepspeed
• 18:1518:15, 25 May 2023 diff hist +23‎ m Deepspeed ‎ →‎Multi-GPU and multi-node jobs with Deepspeed
• 18:1418:14, 25 May 2023 diff hist −33‎ Deepspeed ‎ →‎Multi-GPU and multi-node jobs with Deepspeed
• 18:0518:05, 25 May 2023 diff hist −38‎ m Deepspeed ‎ →‎Multi-GPU and multi-node jobs with Deepspeed
• 17:3417:34, 25 May 2023 diff hist −204‎ m Deepspeed ‎ →‎Multi-GPU and multi-node jobs with Deepspeed
• 15:5515:55, 25 May 2023 diff hist +7,809‎ N Deepspeed ‎ Created page with "DeepSpeed is a deep learning training optimization library, providing the means to train massive billion parameter models at scale. Fully compatible with PyTorch, DeepSpeed features implementations of novel memory-efficient distributed training methods, based on the Zero Redundancy Optimizer (ZeRO) concept. Through the use of ZeRO, DeepSpeed enables distributed storage and computing of different elements of a training task - such as optimizer states, model weights, model..."
• 15:3215:32, 25 May 2023 diff hist +547‎ Huggingface ‎ →‎Accelerate
• 15:1715:17, 25 May 2023 diff hist +4,751‎ Huggingface ‎No edit summary

23 May 2023

• 18:4718:47, 23 May 2023 diff hist +226‎ Huggingface ‎No edit summary