StackOverflow Questions for Tag: accelerate

Amr Samer

Reputation: 41

AttributeError: 'AcceleratorState' object has no attribute 'distributed_type'

huggingfaceaccelerate

Score: 4

Answers: 2

Noah

Reputation: 11

How to fix a bug about Accelerate which is reported 'AcceleratorState' object has no attribute 'distributed_type' by the Pycharm

pytorchhuggingfacefine-tuningaccelerate

Score: 1

Answers: 0

JackS14235

Reputation: 9

How can I efficiently invert a large matrix in Swift using the Accelerate framework?

swiftmatrixaccelerate

Score: 0

Answers: 0

Mohammad Sadoughi

Reputation: 1179

ImportError: Using `low_cpu_mem_usage=True` or a `device_map` requires Accelerate: `pip install accelerate`

huggingface-transformersamazon-sagemaker-studioaccelerate

Score: 19

Answers: 7

Mico Mrkaic

Reputation: 29

GNU Octave on macOS with Accelerate framework

macosoctaveaccelerate

Score: 0

Answers: 0

Pavithran Iyer

Reputation: 412

Using a LAPACK function in C from the Accelerate Package on a M1 Mac

clapackaccelerate-frameworkaccelerate

Score: 1

Answers: 0

xjabr

Reputation: 153

ImportError: Using `bitsandbytes` 8-bit quantization requires Accelerate: `pip install accelerate`

pythonmachine-learninghuggingface-transformersllamaaccelerate

Score: 5

Answers: 6

Enrico Shippole

Reputation: 97

Get attention masks from HF pipelines

nlphuggingface-transformershuggingfacehuggingface-tokenizersaccelerate

Score: 1

Answers: 2

A.A

Reputation: 4151

FSDP in Accelerate for Large-Context LLaMA Training

pythonpytorchhuggingface-transformersllamaaccelerate

Score: 0

Answers: 0

Rakesh Jarupula

Reputation: 21

Cuda failure ‘named symbol not found’ when run on 4 L4 GPUs

python-3.xpytorchmultiprocessingdistributed-computingaccelerate

Score: 0

Answers: 0

Eric Fang

Reputation: 1

Why does using 4 GPUs for distributed inference slow down compared to 2 GPUs?

deep-learningpytorchdistributed-computinghuggingfaceaccelerate

Score: 0

Answers: 0

Isbister

Reputation: 946

How can I lock the global batch size? (DeepSpeed)

pytorchhuggingface-transformersdistributed-computingacceleratedeepspeed

Score: 0

Answers: 0

Rocketq

Reputation: 5791

ImportError: Using the `Trainer` with `PyTorch` requires `accelerate>=0.26.0` but I have version already installed 1.0.1

importpytorchhuggingface-transformersaccelerate

Score: 1

Answers: 2

Deshwal

Reputation: 4162

How does the data splitting actually work in Multi GPU Inference for Accelerate when used in a batched inference setting?

pythonpytorchhuggingface-transformershuggingfaceaccelerate

Score: 2

Answers: 1

Yaoming Xuan

Reputation: 176

What does "gradient_accumulation_steps" do in deepspeed?

acceleratemini-batchdeepspeed

Score: 1

Answers: 0

gazm2k5

Reputation: 500

Unexpected memory usage when training transformer models with LoRA and quantization

pythonhuggingface-transformershuggingfaceaccelerate

Score: 0

Answers: 0

dan

Reputation: 1

Performing a FFT using Accelerate in Swift using a sample rate that is not a power of 2

swiftfftaccelerate

Score: 0

Answers: 1

StaEx_G

Reputation: 13

Training LLM uses unexpected amount of GPU memory

pytorchartificial-intelligencehuggingface-transformerslarge-language-modelaccelerate

Score: 0

Answers: 0

Llmw123

Reputation: 21

How do I free up GPU memory when using Accelerate with Deepspeed

out-of-memoryhuggingface-transformersacceleratedeepspeed

Score: 2

Answers: 0

Aswin Jimmy

Reputation: 21

ImportError: Using `bitsandbytes` 8-bit quantization requires Accelerate

machine-learninghuggingface-transformerssummarizeacceleratellama3

Score: 2

Answers: 0

PreviousPage 1Next

StackOverflow Questions for Tag: accelerate

AttributeError: &#39;AcceleratorState&#39; object has no attribute &#39;distributed_type&#39;

How to fix a bug about Accelerate which is reported &#39;AcceleratorState&#39; object has no attribute &#39;distributed_type&#39; by the Pycharm

How can I efficiently invert a large matrix in Swift using the Accelerate framework?

ImportError: Using `low_cpu_mem_usage=True` or a `device_map` requires Accelerate: `pip install accelerate`

GNU Octave on macOS with Accelerate framework

Using a LAPACK function in C from the Accelerate Package on a M1 Mac

ImportError: Using `bitsandbytes` 8-bit quantization requires Accelerate: `pip install accelerate`

Get attention masks from HF pipelines

FSDP in Accelerate for Large-Context LLaMA Training

Cuda failure ‘named symbol not found’ when run on 4 L4 GPUs

Why does using 4 GPUs for distributed inference slow down compared to 2 GPUs?

How can I lock the global batch size? (DeepSpeed)

ImportError: Using the `Trainer` with `PyTorch` requires `accelerate&gt;=0.26.0` but I have version already installed 1.0.1

How does the data splitting actually work in Multi GPU Inference for Accelerate when used in a batched inference setting?

What does &quot;gradient_accumulation_steps&quot; do in deepspeed?

Unexpected memory usage when training transformer models with LoRA and quantization

Performing a FFT using Accelerate in Swift using a sample rate that is not a power of 2

Training LLM uses unexpected amount of GPU memory

How do I free up GPU memory when using Accelerate with Deepspeed

ImportError: Using `bitsandbytes` 8-bit quantization requires Accelerate

AttributeError: 'AcceleratorState' object has no attribute 'distributed_type'

How to fix a bug about Accelerate which is reported 'AcceleratorState' object has no attribute 'distributed_type' by the Pycharm

ImportError: Using the `Trainer` with `PyTorch` requires `accelerate>=0.26.0` but I have version already installed 1.0.1

What does "gradient_accumulation_steps" do in deepspeed?