qbcompiler.configs.mod_config.ModConfig Class Reference#
|
Mobilint SDK qb Compiler v1.0
MCS002-EN
|
qbcompiler.configs.mod_config.ModConfig Class Reference
Configuration for Minimum Output Difference algorithm. More...
Inheritance diagram for qbcompiler.configs.mod_config.ModConfig:
Public Member Functions | |
| __init__ (self, bool apply, int epochs, int warmup_epochs, numbers.Number lr_act_scale, numbers.Number lr_zeropoint, numbers.Number lr_weight_scale, numbers.Number lr_weight, numbers.Number lr_bias, numbers.Number lr_min_ratio, int batch_size, numbers.Number q_drop, str post_processing, numbers.Number box_conf_thres, numbers.Number box_iou_thres, str loss_type, bool use_outputs, numbers.Number kl_temperature, numbers.Number recon_prob, numbers.Number recon_coeff, numbers.Number lambda_0, numbers.Number lambda_1, numbers.Number lambda_2, List anchors, bool use_xyxy, str save_dir, int seed, List[str] apply_layers, List[str] exclude_layers, str custom_loss_jit_path, str mod_after_layer_name, bool optional=optional) | |
| Initialize the ModConfig. | |
| default_config (cls) | |
| from_dict (cls, dict config_dict) | |
| Create ModConfig from dictionary (JSON structure) | |
| to_dict (self) | |
Public Attributes | |
| apply = apply | |
| epochs = epochs | |
| warmup_epochs = warmup_epochs | |
| lr_act_scale = lr_act_scale | |
| lr_zeropoint = lr_zeropoint | |
| lr_weight_scale = lr_weight_scale | |
| lr_weight = lr_weight | |
| lr_bias = lr_bias | |
| lr_min_ratio = lr_min_ratio | |
| batch_size = batch_size | |
| q_drop = q_drop | |
| post_processing = post_processing | |
| box_conf_thres = box_conf_thres | |
| box_iou_thres = box_iou_thres | |
| loss_type = loss_type | |
| use_outputs = use_outputs | |
| kl_temperature = kl_temperature | |
| recon_prob = recon_prob | |
| recon_coeff = recon_coeff | |
| lambda_0 = lambda_0 | |
| lambda_1 = lambda_1 | |
| lambda_2 = lambda_2 | |
| anchors = anchors | |
| use_xyxy = use_xyxy | |
| save_dir = save_dir | |
| seed = seed | |
| apply_layers = apply_layers | |
| exclude_layers = exclude_layers | |
| custom_loss_jit_path = custom_loss_jit_path | |
| mod_after_layer_name = mod_after_layer_name | |
Static Public Attributes | |
| bool | optional = True |
| dict | DEFAULTS |
| dict | py_to_cpp_name |
| dict | min_max_check |
Detailed Description
Configuration for Minimum Output Difference algorithm.
Defines parameters controlling whether and how Mod is applied during quantization, including layer-level inclusion/exclusion lists.
Definition at line 14 of file mod_config.py.
The documentation for this class was generated from the following file:
Generated by
1.14.0