MCPcopy Create free account

hub / github.com/CERT-Lab/lora-sb / functions

Functions134 in github.com/CERT-Lab/lora-sb

Methodafter_hook
(module: torch.nn.Module, args, output)
utils/offload_utils_for_quant/forward_hook.py:65
Methodafter_hook
(module, grad_input, grad_output)
utils/offload_utils_for_quant/forward_backward_offload.py:178
Methodafter_hook_with_kwargs
(module, args, kwargs, output)
utils/offload_utils_for_quant/forward_hook.py:78
Methodafter_hook_with_kwargs
(module, args, kwargs, output)
utils/offload_utils_for_quant/forward_backward_offload.py:146
Functionclean_numbers
(sample)
instruction_tuning_eval/utils.py:57
Functioncleanup_wandb
()
train_glue.py:53
Functioncompute_metrics
(eval_preds)
utils/compute_utils.py:18
Functioncreate_compute_metrics_function
Create a compute_metrics function with the required parameters.
utils/compute_utils.py:3
Functioncreate_model_tokenizer_cr
(args)
models.py:92
Functionforward_latent
(self, x: torch.Tensor)
utils/latent_utils.py:53
Functiongenerate_and_tokenize_prompt_wrapper
(data_point)
utils/data_utils.py:246
Functionget_delta_weight
Compute the delta weight for the given adapter. Args: adapter (str): The name of the adapter for which the delta weight
utils/latent_utils.py:10
Methodget_full_name_list
Get the module name list of the leaf nodes of the module tree
utils/offload_utils_for_quant/forward_hook.py:95
Functionget_lr_scheduler
(optimizer, num_warmup_steps, num_training_steps)
train_eval.py:139
Functionget_model_memory
Calculates the estimated memory usage of a model in gigabytes. Args: model (torch.nn.Module): The model whose memory usage is to be
utils/offload_utils_for_quant/split.py:4
Methodhook
(module: torch.nn.Module, args)
utils/offload_utils_for_quant/forward_hook.py:14
Methodhook_with_kwargs
(module: torch.nn.Module, args, kwargs)
utils/offload_utils_for_quant/forward_hook.py:25
Functionis_equiv
(str1, str2, verbose=False)
instruction_tuning_eval/utils.py:237
Functionlast_boxed_only
(sample)
instruction_tuning_eval/utils.py:5
Functionmath_equal_process
(param)
instruction_tuning_eval/grader.py:98
Functiononly_until_first_boxed_from_tokens
(string, tokens)
instruction_tuning_eval/utils.py:41
Methodpack
(x: torch.Tensor)
utils/offload_utils_for_quant/graph_hook.py:24
Methodpre_hook
(module: torch.nn.Module, args)
utils/offload_utils_for_quant/forward_hook.py:60
Methodpre_hook
(module, grad_output)
utils/offload_utils_for_quant/forward_backward_offload.py:174
Methodpre_hook_with_kwargs
(module, args, kwargs)
utils/offload_utils_for_quant/forward_hook.py:72
Methodpre_hook_with_kwargs
(module, args, kwargs)
utils/offload_utils_for_quant/forward_backward_offload.py:126
Functionrecord_gradient_hook
(grad)
utils/gradient_utils.py:23
Methodrecord_gradient_hook
(grad)
utils/offload_utils_for_quant/gradient_offload.py:46
Functionsymbolic_equal_process
(a, b, output_queue)
instruction_tuning_eval/grader.py:128
Functiontokenize_function
(examples)
utils/data_utils.py:34
Functiontrain_client
(model, dataloader, optimizer, scheduler, args)
train_eval.py:37
Functiontrain_tokenize_function
(examples, tokenizer, query, response)
utils/data_utils.py:209
Methodunpack
(packed)
utils/offload_utils_for_quant/graph_hook.py:19
Functionupdate_decoder_weights
(target_module, new_weight)
utils/initialization_utils.py:57
← previous101–134 of 134, ranked by callers