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. 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. FAQ This section contains a list of frequently asked questions that will be of interest to many users

Topic	Replies	Views	Activity
About the Training category Training	0	927	December 21, 2020
RuntimeError: [Rank:0] FATAL ERROR :: MODULE:PT_BRIDGE Exception in Lowering thread PyTorch	3	356	March 23, 2025
RuntimeError: Input sizes must be equal when doing loss.backward() during the training of a GNN PyTorch pytorch	3	186	March 20, 2025
Activation checkpointing modules with kwargs in forward PyTorch	1	178	January 19, 2025
AttributeError: module 'habana_frameworks.torch.hpu' has no attribute 'wrap_in_hpu_graph PyTorch	4	231	January 19, 2025
Hccl failure to be connected on two nodes with a simple script PyTorch	1	197	January 2, 2025
Training of torch.nn.embedding failed: loss not decreasing PyTorch pytorch	2	124	January 2, 2025
Transferring kNN results from CPU to HPU breaks back propagation PyTorch	0	101	December 3, 2024
Synapse detected a device critical error that requires a restart. [Compute or dma timeout] PyTorch pytorch	0	182	November 12, 2024
NotImplementedError: Could not run 'aten::_sparse_coo_tensor_with_dims_and_tensors' with arguments from the 'SparseHPU' backend PyTorch	1	314	November 12, 2024
hl_qual resnet50 training failed when load resnet50DecoderConfig.ini running on Gaudi2D Training	0	130	November 12, 2024
GCNConv fails with normalization PyTorch	0	130	November 5, 2024
AttributeError : 'HabanaParameterWrapper' object has no attribute 'change_device_placement' Training pytorch	6	226	October 23, 2024
AttributeError : 'HabanaParameterWrapper' object has no attribute 'change_device_placement' PyTorch pytorch	1	119	September 24, 2024
VITS training got RuntimeError: MKL FFT doesn't support tensors of type: BFloat16 Training	4	324	August 23, 2024
Issue running Llama2 pretraining using megatron deepspeed PyTorch pytorch	2	181	August 1, 2024
PRELU RuntimeError for inputs more than 1 dimension Training	3	169	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	171	July 22, 2024
Problem with training llama-3-70b with deepspeed Training pytorch	1	272	July 18, 2024
Hpu_backend not found on torch.compile PyTorch	2	356	July 11, 2024
Does HPU can use torch.nn.utils.weight_norm or spectral_norm Training	0	147	June 7, 2024
Pytorch complex datatype Training pytorch	1	176	May 28, 2024
Does HPU support complex datatype in torch Training pytorch	1	280	May 28, 2024
Why is there no hello-world level tutorial for using the Gaudi chip? Training	3	474	April 23, 2024
Does Gaudi support CUDA? FAQ models	2	2509	April 23, 2024
RuntimeError: No backend type associated with device type cpu PyTorch	2	1490	April 19, 2024
Gaudi1 HPU doesn't support long? PyTorch pytorch	11	432	April 4, 2024
SyncBatchNorm Error PyTorch	5	427	March 21, 2024
Model.to device faile: "RuntimeError: synStatus=8 [Device not found] Device acquire failed." Training models , pytorch	3	736	March 13, 2024
Synapse detected a device critical error TensorFlow	3	943	December 21, 2023