Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[CVS-91469] Add a deprecated sseg model template to support compatibility #1264

Merged
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Original file line number Diff line number Diff line change
@@ -0,0 +1,15 @@
metric: mDice
search_algorithm: asha
hp_space:
learning_parameters.learning_rate:
param_type: qloguniform
range:
- 0.0002
- 0.005
- 0.0001
learning_parameters.batch_size:
param_type: qloguniform
range:
- 6
- 12
- 2
Original file line number Diff line number Diff line change
@@ -0,0 +1,6 @@
_base_ = [
'../../../submodule/models/segmentation/ocr_litehrnet18_mod2.yaml',
]

load_from = 'https://storage.openvinotoolkit.org/repositories/openvino_training_extensions/models/custom_semantic_segmentation/litehrnet18_imagenet1k_rsc.pth'
fp16 = dict(loss_scale=512.)
Original file line number Diff line number Diff line change
@@ -0,0 +1,52 @@
# Description.
model_template_id: Custom_Semantic_Segmentation_Lite-HRNet-18_OCR
name: Lite-HRNet-18
task_type: SEGMENTATION
task_family: VISION
instantiation: "CLASS"
summary: Class-Incremental Semantic Segmentation with middle-sized architecture which based on the Lite-HRNet backbone for the balance between the fast inference and long training. (deprecated in next version)
application: ~

# Algo backend.
framework: OTESegmentation v0.14.0

# Task implementations.
entrypoints:
base: mpa_tasks.apis.segmentation.SegmentationTrainTask
openvino: segmentation_tasks.apis.segmentation.OpenVINOSegmentationTask
nncf: mpa_tasks.apis.segmentation.SegmentationNNCFTask
base_model_path: ../../../../mmsegmentation/configs/custom-sematic-segmentation/ocr-lite-hrnet-18/template_experimental.yaml

# Capabilities.
capabilities:
- compute_representations

# Hyperparameters.
hyper_parameters:
base_path: ../configuration.yaml
parameter_overrides:
learning_parameters:
batch_size:
default_value: 8
learning_rate:
default_value: 0.001
auto_hpo_state: POSSIBLE
learning_rate_fixed_iters:
default_value: 0
learning_rate_warmup_iters:
default_value: 100
num_iters:
default_value: 300
algo_backend:
train_type:
default_value: Incremental

# Training resources.
max_nodes: 1
training_targets:
- GPU
- CPU

# Stats.
gigaflops: 3.45
size: 4.5