Training

TensorFlow Welcome to the TensorFlow Category. This section is for specific questions related to Training models using the TensorFlow framework. Please refer to the Gaudi Release Notes for the latest version of TensorFlow that is supported. FAQ This section contains a list of frequently asked questions that will be of interest to many users PyTorch Welcome to the PyTorch Category. This section is for specific questions related to Training models using the PyTorch framework. Please refer to the Gaudi Release Notes for the latest version of PyTorch that is supported.

Topic	Replies	Views	Activity
About the Training category Training	0	842	December 21, 2020
VITS training got RuntimeError: MKL FFT doesn't support tensors of type: BFloat16 Training	4	41	August 23, 2024
Issue running Llama2 pretraining using megatron deepspeed PyTorch pytorch	2	46	August 1, 2024
PRELU RuntimeError for inputs more than 1 dimension Training	3	30	July 23, 2024
Problem to train the local dataset using Llama2 Fine-Tuning with Low-Rank Adaptations (LoRA) on Intel® Gaudi®2 AI Accelerator Training models , advisory , tensorflow	3	24	July 22, 2024
Problem with training llama-3-70b with deepspeed Training pytorch	1	40	July 18, 2024
Hpu_backend not found on torch.compile PyTorch	2	62	July 11, 2024
Does HPU can use torch.nn.utils.weight_norm or spectral_norm Training	0	71	June 7, 2024
Pytorch complex datatype Training pytorch	1	84	May 28, 2024
Does HPU support complex datatype in torch Training pytorch	1	89	May 28, 2024
Why is there no hello-world level tutorial for using the Gaudi chip? Training	3	218	April 23, 2024
Does Gaudi support CUDA? FAQ models	2	2086	April 23, 2024
RuntimeError: No backend type associated with device type cpu PyTorch	2	516	April 19, 2024
Gaudi1 HPU doesn't support long? PyTorch pytorch	11	219	April 4, 2024
SyncBatchNorm Error PyTorch	5	218	March 21, 2024
Model.to device faile: "RuntimeError: synStatus=8 [Device not found] Device acquire failed." Training models , pytorch	3	328	March 13, 2024
Synapse detected a device critical error TensorFlow	3	602	December 21, 2023
Trainer killed/Segfault PyTorch	6	454	September 1, 2023
Something similar to CUDA_VISIBLE_DEVICES Training	7	1134	July 20, 2023
libSynapse.so: undefined symbol: synEventMapTensorBase TensorFlow tensorflow	1	376	June 30, 2023
On the steps of integrating habana-horovod with TensorFlow TensorFlow	2	385	June 26, 2023
Multi-node non-mlperf Resnet50 training with Horovod Training	1	495	June 23, 2023
How to use hccl with horovod? Training	1	435	June 13, 2023
Gaudi2 Mlperf v2.1 multi node support Training	1	390	June 13, 2023
Gaudi2 slower compared to A100 Training	10	550	June 7, 2023
ValueError: invalid type: 'torch.hpu.FloatTensor' Training	9	653	June 6, 2023
multi-node training with horovod failing with Synpase error but ports are online Training	1	471	June 6, 2023
Gaudi eval dataset in tfrecord format to get accuracy of run Training	15	442	April 11, 2023
unet2d training crash for 8 gaudis Training pytorch	2	615	March 17, 2023
Advisory: $HOME/.habana_logs can be created without write/execute permissions FAQ advisory	1	747	March 16, 2023