Method_compute_preconditioned_gradient(
self,
grad: Tensor,
preconditioned_dims_selector: tuple[bool, ...],
kronecke
records/track_3_optimization/results/20260513_shampoo_1_4_power/distributed_shampoo/preconditioner/shampoo_preconditioner_list.py:1739
Method_construct_model(
model_linear_layers_dims: tuple[int, ...],
model_dead_layers_dims: tuple[int, ...],
records/track_3_optimization/results/20260513_shampoo_1_4_power/distributed_shampoo/distributor/gpu_tests/shampoo_hybrid_shard_lossless_distributor_test.py:86
Method_construct_model(
model_linear_layers_dims: tuple[int, ...],
model_dead_layers_dims: tuple[int, ...],
records/track_3_optimization/results/20260513_shampoo_1_4_power/distributed_shampoo/distributor/gpu_tests/shampoo_fully_shard_lossless_distributor_test.py:84
Method_convert_float_to_float(
src_list: tuple[Tensor, ...],
target_dtype: torch.dtype,
dest_list: tuple[Tensor, ..
records/track_3_optimization/results/20260513_shampoo_1_4_power/distributed_shampoo/utils/shampoo_quantization.py:258