Add BottomUp model pipeline #52
Conversation
WalkthroughThe changes introduce new predictor classes ( Changes
Sequence Diagram(s)sequenceDiagram
participant User
participant Predictor
participant Model
participant DataPipeline
participant SLEAPData
User ->> Predictor: Initialize from trained models
Predictor ->> Model: Load model
User ->> Predictor: Create data pipeline
Predictor ->> DataPipeline: Initialize pipeline
User ->> Predictor: Run inference
Predictor ->> Model: Predict data
Model ->> Predictor: Return predictions
Predictor ->> SLEAPData: Convert to SLEAP format
SLEAPData ->> User: Return labeled frames
Possibly related issues
Poem
Thank you for using CodeRabbit. We offer it for free to the OSS community and would appreciate your support in helping us grow. If you find it useful, would you consider giving us a shout-out on your favorite social media? TipsChatThere are 3 ways to chat with CodeRabbit:
Note: Be mindful of the bot's finite context window. It's strongly recommended to break down tasks such as reading entire modules into smaller chunks. For a focused discussion, use review comments to chat about specific files and their changes, instead of using the PR comments. CodeRabbit Commands (invoked as PR comments)
Additionally, you can add CodeRabbit Configration File (
|
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #52 +/- ##
==========================================
+ Coverage 96.64% 97.71% +1.07%
==========================================
Files 23 29 +6
Lines 1818 2672 +854
==========================================
+ Hits 1757 2611 +854
Misses 61 61 ☔ View full report in Codecov by Sentry. |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 10
Outside diff range and nitpick comments (23)
tests/inference/test_single_instance.py (1)
1-1: Add a module-level docstring to describe the purpose and contents of this test file.tests/inference/test_bottomup.py (1)
1-1: Add a module-level docstring to describe the purpose and contents of this test file.tests/data/test_confmaps.py (1)
Line range hint
1-1: Add a module-level docstring to describe the purpose and contents of this test file.tests/inference/test_paf_grouping.py (18)
Line range hint
1-1: Add a module-level docstring to describe the purpose and scope of the tests in this file.
Line range hint
28-28: Add a docstring totest_get_connection_candidatesto explain the purpose of the test and its parameters.
Line range hint
44-44: Add a docstring totest_make_line_substo explain the purpose of the test and its parameters.
Line range hint
58-58: Add a docstring totest_get_paf_linesto explain the purpose of the test and its parameters.
Line range hint
78-78: Add a docstring totest_compute_distance_penaltyto explain the purpose of the test and its parameters.
Line range hint
94-94: Add a docstring totest_score_paf_linesto explain the purpose of the test and its parameters.
Line range hint
117-117: Add a docstring totest_score_paf_lines_batchto explain the purpose of the test and its parameters.
Line range hint
153-153: Add a docstring totest_match_candidates_sampleto explain the purpose of the test and its parameters.
Line range hint
179-179: Add a docstring totest_match_candidates_batchto explain the purpose of the test and its parameters.
Line range hint
205-205: Add a docstring totest_toposort_edgesto explain the purpose of the test and its parameters.
Line range hint
245-245: Add a docstring totest_assign_connections_to_instancesto explain the purpose of the test and its parameters.
Line range hint
339-339: Add a docstring totest_make_predicted_instancesto explain the purpose of the test and its parameters.
Line range hint
367-367: Add a docstring totest_group_instances_sampleto explain the purpose of the test and its parameters.
Line range hint
415-415: Add a docstring totest_group_instances_batchto explain the purpose of the test and its parameters.
Line range hint
476-476: Add a docstring totest_paf_scorer_from_configto explain the purpose of the test and its parameters.
Line range hint
487-487: Add a docstring totest_paf_scorer_score_paf_linesto explain the purpose of the test and its parameters.
Line range hint
531-531: Add a docstring totest_paf_scorer_match_candidatesto explain the purpose of the test and its parameters.
Line range hint
569-569: Add a docstring totest_paf_scorer_group_instancesto explain the purpose of the test and its parameters.tests/data/test_pipelines.py (2)
Line range hint
1-1: Add a module-level docstring to describe the purpose and contents of this test module.
460-460: Ensure consistency in the configuration ofconf_map_genacross different tests to avoid discrepancies in expected outcomes.Also applies to: 525-525
Review Details
Configuration used: CodeRabbit UI
Review profile: CHILL
Files selected for processing (27)
- docs/config.md (6 hunks)
- docs/config_bottomup.yaml (1 hunks)
- docs/config_centroid.yaml (4 hunks)
- sleap_nn/architectures/model.py (4 hunks)
- sleap_nn/data/confidence_maps.py (6 hunks)
- sleap_nn/data/edge_maps.py (1 hunks)
- sleap_nn/data/instance_cropping.py (1 hunks)
- sleap_nn/data/pipelines.py (2 hunks)
- sleap_nn/data/providers.py (1 hunks)
- sleap_nn/inference/bottomup.py (1 hunks)
- sleap_nn/inference/paf_grouping.py (4 hunks)
- sleap_nn/inference/predictors.py (15 hunks)
- sleap_nn/inference/single_instance.py (1 hunks)
- sleap_nn/inference/topdown.py (1 hunks)
- sleap_nn/training/model_trainer.py (6 hunks)
- tests/assets/minimal_instance_bottomup/initial_config.yaml (1 hunks)
- tests/assets/minimal_instance_bottomup/training_config.yaml (1 hunks)
- tests/data/test_confmaps.py (2 hunks)
- tests/data/test_instance_cropping.py (1 hunks)
- tests/data/test_pipelines.py (4 hunks)
- tests/fixtures/datasets.py (1 hunks)
- tests/inference/test_bottomup.py (1 hunks)
- tests/inference/test_paf_grouping.py (1 hunks)
- tests/inference/test_predictors.py (1 hunks)
- tests/inference/test_single_instance.py (1 hunks)
- tests/inference/test_topdown.py (1 hunks)
- tests/training/test_model_trainer.py (3 hunks)
Files skipped from review due to trivial changes (1)
- sleap_nn/data/edge_maps.py
Additional Context Used
LanguageTool (51)
docs/config.md (51)
Near line 6: Loose punctuation mark.
Context: ... four main sections: - 1.data_config: Creating a data pipeline. - 2. `model_...
Near line 8: Loose punctuation mark.
Context: ...ng a data pipeline. - 2.model_config: Initialise the sleap-nn backbone and he...
Near line 10: Loose punctuation mark.
Context: ... and head models. - 3.trainer_config: Hyperparameters required to train the m...
Near line 12: Loose punctuation mark.
Context: ...with Lightning. - 4.inference_config: Inference related configs. Note:...
Near line 16: Loose punctuation mark.
Context: ... forval_data_loader. -data_config: -provider: (str) Provider class...
Near line 19: Loose punctuation mark.
Context: ...psPipeline" or "BottomUp". -train: -labels_path: (str) Path to ...
Near line 21: Possible missing article found.
Context: ...he image has 3 channels (RGB image). If input has only one channel when this ...
Near line 22: Possible missing article found.
Context: ... is set toTrue, then the images from single-channel is replicated along the...
Near line 23: Possible missing article found.
Context: ...s replicated along the channel axis. If input has three channels and this is ...
Near line 24: Possible missing article found.
Context: ... to False, then we convert the image to grayscale (single-channel) image. ...
Near line 31: Loose punctuation mark.
Context: ...e same factor. -preprocessing: -anchor_ind: (int) Index...
Near line 32: Possible missing comma found.
Context: ...can significantly improve topdown model accuracy as they benefit from a consistent geome...
Near line 34: Possible missing comma found.
Context: ...space. Larger values are easier to learn but are less precise with respect to the pe...
Near line 34: ‘with respect to’ might be wordy. Consider a shorter alternative.
Context: ...re easier to learn but are less precise with respect to the peak coordinate. This spread is in ...
Near line 36: Loose punctuation mark.
Context: ...n. -augmentation_config: -random crop`: (Dict[...
Near line 64: Loose punctuation mark.
Context: ... totrainstructure) -model_config: -init_weight: (str) model weigh...
Near line 67: Loose punctuation mark.
Context: ...win_B_Weights"]. -backbone_config: -backbone_type: (str) Backbo...
Near line 78: This phrase might be redundant. Consider either removing or replacing the adjective ‘additional’.
Context: ... -middle_block: (bool) If True, add an additional block at the end of the encoder. default: Tru...
Near line 80: Possible missing comma found.
Context: ... for upsampling. Interpolation is faster but transposed convolutions may...
Near line 93: This phrase might be redundant. Consider either removing or replacing the adjective ‘additional’.
Context: ... -middle_block: (bool) If True, add an additional block at the end of the encoder. default: Tru...
Near line 95: Possible missing comma found.
Context: ... for upsampling. Interpolation is faster but transposed convolutions may...
Near line 121: Possible missing comma found.
Context: ... for upsampling. Interpolation is faster but transposed convolutions may...
Near line 125: Possible missing comma found.
Context: ... for upsampling. Interpolation is faster but transposed convolutions may...
Near line 130: Loose punctuation mark.
Context: .... Default: "tiny". -arch: Dictionary of embed dimension, depths a...
Near line 134: Loose punctuation mark.
Context: .... Default: "tiny". -arch: Dictionary of embed dimension, depths a...
Near line 147: Possible missing comma found.
Context: ... for upsampling. Interpolation is faster but transposed convolutions may...
Near line 150: Possible missing comma found.
Context: ...List[dict]) List of heads in the model. For eg, BottomUp model has both 'MultiInsta...
Near line 156: Possible missing comma found.
Context: ...can significantly improve topdown model accuracy as they benefit from a consistent geome...
Near line 157: Possible missing comma found.
Context: ...space. Larger values are easier to learn but are less precise with respect to the pe...
Near line 157: ‘with respect to’ might be wordy. Consider a shorter alternative.
Context: ...re easier to learn but are less precise with respect to the peak coordinate. This spread is in ...
Near line 160: Loose punctuation mark.
Context: ...ic output in multi-head models. ] -trainer_config: - `train_data...
Near line 168: Possible missing article found.
Context: ...he batch size. If False and the size of dataset is not divisible by the batch size, the...
Near line 172: Possible typo: you repeated a word
Context: ...ease note that the monitors are checked every every_n_epochs epochs. if save_top_k >= 2 and...
Near line 172: Possible typo: you repeated a word
Context: ... the monitors are checked every every_n_epochs epochs. if save_top_k >= 2 and the callback is...
Near line 202: Possible missing comma found.
Context: ...onitored has stopped decreasing; in max mode it will be reduced when the quantity mo...
Near line 206: Possible missing comma found.
Context: ...tience`: (int) Number of epochs with no improvement after which learning rate will be reduc...
Near line 208: Possible missing comma found.
Context: ...arning rate of all param groups or each group respectively. Default: 0. - `inferen...
Near line 210: Loose punctuation mark.
Context: ...ely. Default: 0. -inference_config: -device: (str) Device on which t...
Near line 212: Loose punctuation mark.
Context: ... "ideep", "hip", "msnpu"). -data: -path: (str) Path to.slp...
Near line 219: Possible missing article found.
Context: ...he image has 3 channels (RGB image). If input has only one channel when this ...
Near line 220: Possible missing article found.
Context: ... is set toTrue, then the images from single-channel is replicated along the...
Near line 221: Possible missing article found.
Context: ...s replicated along the channel axis. If input has three channels and this is ...
Near line 222: Possible missing article found.
Context: ... to False, then we convert the image to grayscale (single-channel) image. ...
Near line 232: Loose punctuation mark.
Context: ... the default. -preprocessing: -anchor_ind: (int) Inde...
Near line 233: Possible missing comma found.
Context: ...can significantly improve topdown model accuracy as they benefit from a consistent geome...
Near line 237: Loose punctuation mark.
Context: ...the input image. -peak_threshold:floatbetween 0 and 1. Minimum confid...
Near line 238: Loose punctuation mark.
Context: ... be ignored. -integral_refinement: IfNone, returns the grid-aligned pea...
Near line 239: Loose punctuation mark.
Context: ... regression. -integral_patch_size: Size of patches to crop around each rou...
Near line 240: Loose punctuation mark.
Context: ... integer scalar. -return_confmaps: IfTrue, predicted confidence maps wi...
Near line 241: Loose punctuation mark.
Context: ... values and points. -return_pafs: IfTrue, predicted part affinity fiel...
Near line 242: Loose punctuation mark.
Context: ...es and points. -return_paf_graph: IfTrue, the part affinity field grap...
Ruff (28)
tests/data/test_confmaps.py (1)
1-1: Missing docstring in public module
tests/data/test_instance_cropping.py (2)
1-1: Missing docstring in public module
10-10: Missing docstring in public function
tests/data/test_pipelines.py (1)
1-1: Missing docstring in public module
tests/inference/test_bottomup.py (1)
1-1: Missing docstring in public module
tests/inference/test_paf_grouping.py (18)
1-1: Missing docstring in public module
28-28: Missing docstring in public function
44-44: Missing docstring in public function
58-58: Missing docstring in public function
78-78: Missing docstring in public function
94-94: Missing docstring in public function
117-117: Missing docstring in public function
153-153: Missing docstring in public function
179-179: Missing docstring in public function
205-205: Missing docstring in public function
245-245: Missing docstring in public function
339-339: Missing docstring in public function
367-367: Missing docstring in public function
415-415: Missing docstring in public function
476-476: Missing docstring in public function
487-487: Missing docstring in public function
531-531: Missing docstring in public function
569-569: Missing docstring in public function
tests/inference/test_predictors.py (1)
1-1: Missing docstring in public module
tests/inference/test_single_instance.py (1)
1-1: Missing docstring in public module
tests/inference/test_topdown.py (1)
1-1: Missing docstring in public module
tests/training/test_model_trainer.py (2)
71-71: Missing docstring in public function
271-271: Missing docstring in public function
Markdownlint (193)
docs/config.md (193)
17: Expected: 2; Actual: 4
Unordered list indentation
18: Expected: 2; Actual: 4
Unordered list indentation
19: Expected: 2; Actual: 4
Unordered list indentation
20: Expected: 4; Actual: 8
Unordered list indentation
21: Expected: 4; Actual: 8
Unordered list indentation
26: Expected: 4; Actual: 8
Unordered list indentation
28: Expected: 4; Actual: 8
Unordered list indentation
30: Expected: 4; Actual: 8
Unordered list indentation
31: Expected: 4; Actual: 8
Unordered list indentation
32: Expected: 6; Actual: 12
Unordered list indentation
33: Expected: 6; Actual: 12
Unordered list indentation
34: Expected: 6; Actual: 12
Unordered list indentation
35: Expected: 6; Actual: 12
Unordered list indentation
36: Expected: 6; Actual: 12
Unordered list indentation
37: Expected: 8; Actual: 16
Unordered list indentation
38: Expected: 8; Actual: 16
Unordered list indentation
39: Expected: 8; Actual: 16
Unordered list indentation
40: Expected: 10; Actual: 20
Unordered list indentation
41: Expected: 12; Actual: 24
Unordered list indentation
42: Expected: 12; Actual: 24
Unordered list indentation
43: Expected: 12; Actual: 24
Unordered list indentation
44: Expected: 12; Actual: 24
Unordered list indentation
45: Expected: 12; Actual: 24
Unordered list indentation
46: Expected: 12; Actual: 24
Unordered list indentation
47: Expected: 12; Actual: 24
Unordered list indentation
48: Expected: 12; Actual: 24
Unordered list indentation
49: Expected: 12; Actual: 24
Unordered list indentation
50: Expected: 10; Actual: 20
Unordered list indentation
51: Expected: 12; Actual: 24
Unordered list indentation
52: Expected: 12; Actual: 24
Unordered list indentation
54: Expected: 12; Actual: 24
Unordered list indentation
55: Expected: 12; Actual: 24
Unordered list indentation
56: Expected: 12; Actual: 24
Unordered list indentation
57: Expected: 12; Actual: 24
Unordered list indentation
58: Expected: 12; Actual: 24
Unordered list indentation
59: Expected: 12; Actual: 24
Unordered list indentation
60: Expected: 12; Actual: 24
Unordered list indentation
61: Expected: 12; Actual: 24
Unordered list indentation
62: Expected: 2; Actual: 4
Unordered list indentation
65: Expected: 2; Actual: 4
Unordered list indentation
66: Expected: 2; Actual: 4
Unordered list indentation
67: Expected: 2; Actual: 4
Unordered list indentation
68: Expected: 4; Actual: 8
Unordered list indentation
69: Expected: 4; Actual: 8
Unordered list indentation
70: Expected: 6; Actual: 12
Unordered list indentation
71: Expected: 6; Actual: 12
Unordered list indentation
72: Expected: 6; Actual: 12
Unordered list indentation
73: Expected: 6; Actual: 12
Unordered list indentation
74: Expected: 6; Actual: 12
Unordered list indentation
76: Expected: 6; Actual: 12
Unordered list indentation
78: Expected: 6; Actual: 12
Unordered list indentation
79: Expected: 6; Actual: 12
Unordered list indentation
83: Expected: 6; Actual: 12
Unordered list indentation
84: Expected: 6; Actual: 12
Unordered list indentation
88: Expected: 6; Actual: 12
Unordered list indentation
89: Expected: 6; Actual: 12
Unordered list indentation
91: Expected: 6; Actual: 12
Unordered list indentation
93: Expected: 6; Actual: 12
Unordered list indentation
94: Expected: 6; Actual: 12
Unordered list indentation
98: Expected: 6; Actual: 12
Unordered list indentation
99: Expected: 6; Actual: 12
Unordered list indentation
103: Expected: 6; Actual: 12
Unordered list indentation
104: Expected: 6; Actual: 12
Unordered list indentation
105: Expected: 4; Actual: 8
Unordered list indentation
106: Expected: 6; Actual: 12
Unordered list indentation
107: Expected: 6; Actual: 12
Unordered list indentation
108: Expected: 8; Actual: 16
Unordered list indentation
109: Expected: 8; Actual: 16
Unordered list indentation
110: Expected: 6; Actual: 12
Unordered list indentation
111: Expected: 6; Actual: 12
Unordered list indentation
112: Expected: 6; Actual: 12
Unordered list indentation
113: Expected: 6; Actual: 12
Unordered list indentation
114: Expected: 6; Actual: 12
Unordered list indentation
115: Expected: 6; Actual: 12
Unordered list indentation
116: Expected: 6; Actual: 12
Unordered list indentation
117: Expected: 6; Actual: 12
Unordered list indentation
118: Expected: 6; Actual: 12
Unordered list indentation
119: Expected: 6; Actual: 12
Unordered list indentation
120: Expected: 6; Actual: 12
Unordered list indentation
124: Expected: 6; Actual: 12
Unordered list indentation
128: Expected: 4; Actual: 8
Unordered list indentation
129: Expected: 6; Actual: 12
Unordered list indentation
130: Expected: 6; Actual: 12
Unordered list indentation
133: Expected: 6; Actual: 12
Unordered list indentation
134: Expected: 6; Actual: 12
Unordered list indentation
137: Expected: 6; Actual: 12
Unordered list indentation
138: Expected: 6; Actual: 12
Unordered list indentation
139: Expected: 6; Actual: 12
Unordered list indentation
140: Expected: 6; Actual: 12
Unordered list indentation
141: Expected: 6; Actual: 12
Unordered list indentation
142: Expected: 6; Actual: 12
Unordered list indentation
143: Expected: 6; Actual: 12
Unordered list indentation
144: Expected: 6; Actual: 12
Unordered list indentation
145: Expected: 6; Actual: 12
Unordered list indentation
146: Expected: 6; Actual: 12
Unordered list indentation
150: Expected: 2; Actual: 4
Unordered list indentation
163: Expected: 2; Actual: 4
Unordered list indentation
164: Expected: 4; Actual: 8
Unordered list indentation
165: Expected: 4; Actual: 8
Unordered list indentation
166: Expected: 4; Actual: 8
Unordered list indentation
167: Expected: 4; Actual: 8
Unordered list indentation
168: Expected: 4; Actual: 8
Unordered list indentation
169: Expected: 4; Actual: 8
Unordered list indentation
170: Expected: 2; Actual: 4
Unordered list indentation
171: Expected: 2; Actual: 4
Unordered list indentation
172: Expected: 4; Actual: 8
Unordered list indentation
173: Expected: 4; Actual: 8
Unordered list indentation
174: Expected: 4; Actual: 8
Unordered list indentation
175: Expected: 4; Actual: 8
Unordered list indentation
176: Expected: 4; Actual: 8
Unordered list indentation
177: Expected: 2; Actual: 4
Unordered list indentation
178: Expected: 4; Actual: 8
Unordered list indentation
179: Expected: 4; Actual: 8
Unordered list indentation
180: Expected: 4; Actual: 8
Unordered list indentation
181: Expected: 2; Actual: 4
Unordered list indentation
182: Expected: 2; Actual: 4
Unordered list indentation
183: Expected: 2; Actual: 4
Unordered list indentation
184: Expected: 2; Actual: 4
Unordered list indentation
185: Expected: 2; Actual: 4
Unordered list indentation
186: Expected: 2; Actual: 4
Unordered list indentation
187: Expected: 2; Actual: 4
Unordered list indentation
188: Expected: 2; Actual: 4
Unordered list indentation
189: Expected: 2; Actual: 4
Unordered list indentation
190: Expected: 2; Actual: 4
Unordered list indentation
191: Expected: 4; Actual: 8
Unordered list indentation
192: Expected: 4; Actual: 8
Unordered list indentation
193: Expected: 4; Actual: 8
Unordered list indentation
194: Expected: 4; Actual: 8
Unordered list indentation
195: Expected: 4; Actual: 8
Unordered list indentation
196: Expected: 4; Actual: 8
Unordered list indentation
197: Expected: 2; Actual: 4
Unordered list indentation
198: Expected: 2; Actual: 4
Unordered list indentation
199: Expected: 4; Actual: 8
Unordered list indentation
200: Expected: 4; Actual: 8
Unordered list indentation
201: Expected: 2; Actual: 4
Unordered list indentation
202: Expected: 4; Actual: 8
Unordered list indentation
203: Expected: 4; Actual: 8
Unordered list indentation
204: Expected: 4; Actual: 8
Unordered list indentation
205: Expected: 4; Actual: 8
Unordered list indentation
206: Expected: 4; Actual: 8
Unordered list indentation
207: Expected: 4; Actual: 8
Unordered list indentation
208: Expected: 4; Actual: 8
Unordered list indentation
211: Expected: 2; Actual: 4
Unordered list indentation
212: Expected: 2; Actual: 4
Unordered list indentation
213: Expected: 4; Actual: 8
Unordered list indentation
214: Expected: 4; Actual: 8
Unordered list indentation
216: Expected: 4; Actual: 8
Unordered list indentation
218: Expected: 4; Actual: 8
Unordered list indentation
219: Expected: 4; Actual: 8
Unordered list indentation
224: Expected: 4; Actual: 8
Unordered list indentation
225: Expected: 4; Actual: 8
Unordered list indentation
226: Expected: 4; Actual: 8
Unordered list indentation
227: Expected: 6; Actual: 12
Unordered list indentation
228: Expected: 6; Actual: 12
Unordered list indentation
229: Expected: 6; Actual: 12
Unordered list indentation
230: Expected: 6; Actual: 12
Unordered list indentation
232: Expected: 4; Actual: 8
Unordered list indentation
233: Expected: 6; Actual: 12
Unordered list indentation
234: Expected: 6; Actual: 12
Unordered list indentation
235: Expected: 6; Actual: 12
Unordered list indentation
236: Expected: 6; Actual: 12
Unordered list indentation
237: Expected: 2; Actual: 4
Unordered list indentation
238: Expected: 2; Actual: 4
Unordered list indentation
239: Expected: 2; Actual: 4
Unordered list indentation
240: Expected: 2; Actual: 4
Unordered list indentation
241: Expected: 2; Actual: 4
Unordered list indentation
242: Expected: 2; Actual: 4
Unordered list indentation
16: Expected: 0 or 2; Actual: 1
Trailing spaces
33: Expected: 0 or 2; Actual: 1
Trailing spaces
35: Expected: 0 or 2; Actual: 1
Trailing spaces
39: Expected: 0 or 2; Actual: 1
Trailing spaces
40: Expected: 0 or 2; Actual: 1
Trailing spaces
64: Expected: 0 or 2; Actual: 1
Trailing spaces
110: Expected: 0 or 2; Actual: 1
Trailing spaces
112: Expected: 0 or 2; Actual: 1
Trailing spaces
129: Expected: 0 or 2; Actual: 1
Trailing spaces
133: Expected: 0 or 2; Actual: 1
Trailing spaces
155: Expected: 0 or 2; Actual: 1
Trailing spaces
162: Expected: 0 or 2; Actual: 1
Trailing spaces
188: Expected: 0 or 2; Actual: 1
Trailing spaces
211: Expected: 0 or 2; Actual: 1
Trailing spaces
223: Expected: 0 or 2; Actual: 1
Trailing spaces
232: Expected: 0 or 2; Actual: 1
Trailing spaces
236: Expected: 0 or 2; Actual: 1
Trailing spaces
240: Expected: 0 or 2; Actual: 1
Trailing spaces
241: Expected: 0 or 2; Actual: 1
Trailing spaces
54: null
Spaces inside emphasis markers
54: null
Spaces inside emphasis markers
54: null
Spaces inside emphasis markers
54: null
Spaces inside emphasis markers
204: null
Spaces inside emphasis markers
204: null
Spaces inside emphasis markers
242: null
Files should end with a single newline character
Additional comments not posted (43)
tests/data/test_instance_cropping.py (1)
29-29: Setmax_instancesto control the number of instances processed. Consider adding a comment explaining why the limit is set to 3.docs/config_centroid.yaml (2)
8-8: The change in scale from 1.0 to 0.5 should be verified to ensure it aligns with the intended data preprocessing adjustments for the centroid model.Verification successful
The initial script did not yield any results, which suggests that the scale parameter set to 0.5 might not be present in other configuration files. To ensure thorough verification, we need to check for the scale parameter in other configurations and model files.
The change in the
scaleparameter from 1.0 to 0.5 indocs/config_centroid.yamlis consistent with the intended data preprocessing adjustments for the centroid model. The verification script confirms that this change is isolated to the specified configuration file and does not conflict with other configurations or model files.
docs/config_centroid.yaml: Thescaleparameter is set to 0.5.This change appears to be intentional and does not introduce inconsistencies with other parts of the codebase.
Scripts Executed
The following scripts were executed for the analysis:
Script:
#!/bin/bash # Description: Verify the scale parameter change in related configuration files and model usage. fd "config_*.yaml" | xargs rg "scale: 0.5"Length of output: 44
Script:
#!/bin/bash # Description: Verify the scale parameter change in all YAML configuration files and Python model files. fd -e yaml -e py | xargs rg "scale:"Length of output: 6435
229-229: Adjusting the scale parameter in the inference configuration is critical. Confirm that this change aligns with the expected model input dimensions and preprocessing during inference.Verification successful
The search results indicate that the
scaleparameter is used in various parts of the codebase, including inference scripts and configuration files. The presence ofscale: 0.5indocs/config_centroid.yamlaligns with other instances wherescaleis set to different values, suggesting that the parameter is indeed used for resizing images during inference.Given this context, it appears that the change to
scale: 0.5in the inference configuration is consistent with the model's expected input dimensions and preprocessing steps.
docs/config_centroid.yaml:scale: 0.5sleap_nn/inference/topdown.py:input_scale: float = 1.0,andself.input_scale = input_scalesleap_nn/inference/single_instance.py:input_scale: float = 1.0,andself.input_scale = input_scalesleap_nn/inference/predictors.py:input_scale=self.centroid_config.inference_config.data.scaleThese references confirm that the
scaleparameter is used to resize images before inference, and setting it to0.5should be valid.Scripts Executed
The following scripts were executed for the analysis:
Script:
#!/bin/bash # Description: Confirm the scale parameter adjustment in the inference configuration aligns with model expectations. fd "config_*.yaml" | xargs rg "scale: 0.5"Length of output: 44
Script:
#!/bin/bash # Description: Search for the usage of the `scale` parameter in the codebase to verify its impact and alignment with model requirements. rg 'scale' --glob '*.py' --glob '*.yaml'Length of output: 21430
tests/assets/minimal_instance_bottomup/initial_config.yaml (1)
1-238: This new configuration file for the BottomUp model appears comprehensive and well-structured, covering essential aspects like data handling, preprocessing, and model settings. Ensure all parameters align with the BottomUp model's requirements and intended use cases.tests/assets/minimal_instance_bottomup/training_config.yaml (1)
1-249: This training configuration file for the BottomUp model is well-prepared, covering training data loaders, model checkpoints, and early stopping settings. Verify that all configurations are optimized for the BottomUp model's training and align with the model's performance expectations.sleap_nn/architectures/model.py (1)
56-56: The modification in theget_backbonefunction to ensure correct application of the backbone configuration is crucial for model integrity and performance.docs/config_bottomup.yaml (1)
1-270: The configuration file for the BottomUp model is comprehensive, covering all necessary settings for data handling, model architecture, and inference. Ensure that all parameters are correctly set to optimize the BottomUp model's performance.sleap_nn/data/providers.py (1)
71-75: The addition of theedge_idxsproperty is a valuable enhancement for the BottomUp model, facilitating graph-based operations by providing easy access to edge indices.sleap_nn/inference/bottomup.py (4)
10-50: The classBottomUpInferenceModelis well-documented and structured. The attributes are clearly defined and relevant to the bottom-up inference process.
52-78: Initialization ofBottomUpInferenceModelis robust, with clear default values and type annotations. Good use of optional parameters for flexibility.
80-101: The method_generate_cms_peakseffectively handles peak detection and adjustment for stride and scale. The use of nested tensors for output is appropriate for handling structured data.
103-161: Theforwardmethod is comprehensive, handling the prediction of confidence maps and the inference of peak coordinates. The method efficiently integrates various components like peak detection and PAF scoring.sleap_nn/data/confidence_maps.py (3)
Line range hint
3-36: The functionmake_confmapsis well-documented with clear parameter descriptions and return type. The implementation uses efficient tensor operations for generating confidence maps.
Line range hint
55-79: The functionmake_multi_confmapsextends the functionality to handle multiple instances. The use oftorch.maximumto combine confidence maps is a good choice for this context.
129-129: The use ofunsqueezein theMultiConfidenceMapGeneratorclass is appropriate for handling tensor dimensions correctly. This ensures that the data is correctly formatted for subsequent operations.tests/fixtures/datasets.py (1)
33-37: The fixtureminimal_instance_bottomup_ckptcorrectly sets up a path for the BottomUp model checkpoint. This is essential for testing the BottomUp model functionality.sleap_nn/data/pipelines.py (1)
293-381: TheBottomUpPipelineclass is well-implemented with clear documentation and a structured approach to building the training pipeline. The integration ofPartAffinityFieldsGeneratoris particularly noteworthy for handling part affinity fields in the bottom-up model.tests/training/test_model_trainer.py (3)
12-17: The import changes correctly include the newBottomUpModelclass, aligning with the PR's objective to add BottomUp model pipeline functionalities.
226-249: The configuration setup for the BottomUp model within the test function is correctly implemented, aligning with the expected parameters for preprocessing and model head configurations.
258-269: The model initialization and training steps are correctly implemented, with appropriate assertions to check the model's output shape and global step.sleap_nn/inference/topdown.py (3)
233-311: LGTM! Ensure all function calls match the new signature.
16-81: LGTM! Ensure all function calls match the new signature.
313-420: LGTM! Ensure all function calls match the new signature.Verification successful
The instances of
FindInstancePeaksin the codebase match the new constructor signature. Here are the relevant locations:
tests/inference/test_topdown.py:
- Line 60
- Line 217
- Line 234
tests/inference/test_inference.py:
- Line 69
- Line 405
- Line 422
sleap_nn/inference/predictors.py:
- Line 318
These instances correctly use the new parameters for
FindInstancePeaks.Scripts Executed
The following scripts were executed for the analysis:
Script:
#!/bin/bash # Description: Verify all function calls to `FindInstancePeaks` match the new signature. # Test: Search for the function usage. Expect: Only occurances of the new signature. rg --type python $'FindInstancePeaks'Length of output: 67
Script:
#!/bin/bash # Description: Verify all function calls to `FindInstancePeaks` match the new signature. # Test: Search for the function usage. Expect: Only occurrences of the new signature. rg --type py 'FindInstancePeaks'Length of output: 3887
Script:
#!/bin/bash # Description: Extract lines where `FindInstancePeaks` is instantiated to verify the new signature. # Extract lines from test files and other relevant files. rg --type py 'FindInstancePeaks\(' -nLength of output: 680
Script:
#!/bin/bash # Description: Extract relevant lines where `FindInstancePeaks` is instantiated to verify the new signature. # Extract lines from test files and other relevant files. rg --type py -A 5 -B 5 'FindInstancePeaks\('Length of output: 6124
sleap_nn/training/model_trainer.py (5)
303-311: LGTM! Ensure all function calls match the new signature.
602-681: LGTM! Ensure all function calls match the new signature.
602-681: LGTM! Ensure all function calls match the new signature.
602-681: LGTM! Ensure all function calls match the new signature.
Line range hint
17-681: LGTM! Ensure all function calls match the new signature.Verification successful
The search results show multiple instances of
ModelTrainerbeing used in the test files. Each instance appears to be using the new signature correctly, as they all pass aconfigobject to theModelTrainerconstructor.
tests/training/test_model_trainer.pytests/inference/test_topdown.pytests/inference/test_bottomup.pytests/inference/test_inference.pyThese instances confirm that the
ModelTrainerclass is being used correctly with the new signature in the test files.Scripts Executed
The following scripts were executed for the analysis:
Script:
#!/bin/bash # Description: Verify all function calls to `ModelTrainer` match the new signature. # Test: Search for the function usage. Expect: Only occurances of the new signature. rg --type python $'ModelTrainer'Length of output: 62
Script:
#!/bin/bash # Description: Verify all function calls to `ModelTrainer` match the new signature. # Test: Search for the function usage. Expect: Only occurrences of the new signature. rg 'ModelTrainer'Length of output: 2107
Script:
#!/bin/bash # Description: Display lines around `ModelTrainer` usage in test files to verify correct usage. # Test: Display lines around `ModelTrainer` usage in test files. rg 'ModelTrainer' -A 5 -B 5 tests/Length of output: 16793
docs/config.md (4)
18-18: Addition of "BottomUp" pipeline option.This addition aligns with the PR's objective to introduce a new model pipeline for BottomUp models. Ensure that the implementation details of this pipeline are well-documented elsewhere in the codebase.
35-35: Introduction ofpafs_genconfiguration for the BottomUp model.This configuration is crucial for defining the generation of Part Affinity Fields (PAFs), which are essential for the BottomUp model's functionality. It's good to see that the configuration mirrors the existing
conf_map_gen, maintaining consistency in the configuration structure.
150-160: Addition of new head types inhead_configs.The inclusion of 'PartAffinityFieldsHead' and 'CenteredInstanceConfmapsHead' is a significant enhancement for the BottomUp model. This change supports the PR's goal of expanding the model's capabilities. Ensure that these new head types are properly integrated and tested within the model's architecture.
236-242: Addition of new inference configuration options.The new options
return_confmaps,return_pafs, andreturn_paf_graphprovide flexibility in the output of the inference process, which is beneficial for detailed analysis and debugging. It's important to ensure that these options are handled correctly in the inference code to avoid any potential performance overhead.tests/data/test_pipelines.py (4)
15-15: The addition ofBottomUpPipelineto the import list aligns with the new features introduced in the PR.
584-584: The introduction of theBottomUpPipelinetest function is crucial for validating the new pipeline's integration and functionality.
584-851: The comprehensive testing of theBottomUpPipelineacross various configurations ensures robustness and functionality. The tests cover different scenarios including scaling, padding, and random cropping, which are essential for validating the pipeline's behavior under different conditions.
584-851: The tests forBottomUpPipelineare well-designed to cover a variety of scenarios, ensuring that the pipeline behaves as expected under different configurations. This is crucial for the reliability of the new pipeline.sleap_nn/inference/predictors.py (4)
854-884: Introduction of theBottomUpPredictorclass.This class encapsulates the functionality for the BottomUp model prediction, aligning with the PR's objectives to enhance bottom-up processing capabilities.
896-956: Initialization logic in_initialize_inference_model.The method correctly sets up the inference model with necessary configurations and scorers, ensuring that the model is ready for inference tasks.
1001-1081: Setup of data loading pipeline inmake_pipeline.The method effectively handles different data providers and sets up the data pipeline accordingly. It is well-structured to accommodate both
LabelsReaderandVideoReader.
1083-1197: Conversion of inference results in_make_labeled_frames_from_generator.The method accurately converts raw inference results into structured data, facilitating further analysis and visualization. The logic is clear and aligns with the framework's data handling standards.
sleap_nn/inference/paf_grouping.py (3)
974-981: Ensure proper handling of device-specific tensors before conversion to NumPy.This conversion is necessary for operations not supported directly in PyTorch, and the handling of device transfer is correctly implemented.
176-176: Ensure tensor operations are performed on the correct device.Also applies to: 178-180
Verification successful
Ensure tensor operations are performed on the correct device.
- The
make_line_subsfunction consistently moves tensors to the device of the input tensors (peaks_sample).Scripts Executed
The following scripts were executed for the analysis:
Script:
#!/bin/bash # Description: Verify that tensor operations in `make_line_subs` are consistently using the device of input tensors. # Test: Search for device-specific operations in `make_line_subs`. Expect: All operations should use the device of input tensors. rg --type python 'make_line_subs' sleap_nn/inference/paf_grouping.pyLength of output: 99
Script:
#!/bin/bash # Description: Verify that tensor operations in `make_line_subs` are consistently using the device of input tensors. # Test: Search for the definition of `make_line_subs` and inspect its device-specific operations. rg 'def make_line_subs' -A 20 sleap_nn/inference/paf_grouping.pyLength of output: 1251
Script:
#!/bin/bash # Description: Verify that tensor operations in `make_line_subs` are consistently using the device of input tensors. # Test: Search for the definition of `make_line_subs` and capture more lines to inspect its device-specific operations. rg 'def make_line_subs' -A 50 sleap_nn/inference/paf_grouping.pyLength of output: 2745
Script:
#!/bin/bash # Description: Verify that tensor operations in `make_line_subs` are consistently using the device of input tensors. # Test: Search for the definition of `make_line_subs` and capture additional lines to inspect its device-specific operations. rg 'def make_line_subs' -A 100 sleap_nn/inference/paf_grouping.pyLength of output: 4225
445-447: Verify the calculation ofmax_edge_lengthfor accuracy.Verification successful
The calculation of
max_edge_lengthin thescore_paf_lines_batchfunction appears to be accurate. It correctly multiplies themax_edge_length_ratioby the maximum dimension ofpafsand thepafs_stride, which logically computes the maximum edge length based on the provided parameters.
max_edge_length_ratio: A scalar that represents the maximum expected length of a connected pair of points.max(pafs.shape[-1], pafs.shape[-2], pafs.shape[-3]): The maximum dimension of thepafstensor.pafs_stride: The stride of thepafs.This calculation ensures that the
max_edge_lengthis appropriately scaled according to the dimensions and stride of thepafs.Scripts Executed
The following scripts were executed for the analysis:
Script:
#!/bin/bash # Description: Verify the calculation of `max_edge_length` in `score_paf_lines_batch`. # Test: Search for usage of `max_edge_length` in `score_paf_lines_batch`. Expect: Correct calculation based on image dimensions and stride. rg --type python 'max_edge_length' sleap_nn/inference/paf_grouping.pyLength of output: 100
Script:
#!/bin/bash # Description: Verify the calculation of `max_edge_length` in `score_paf_lines_batch`. # Test: Search for usage of `max_edge_length` in `score_paf_lines_batch`. Expect: Correct calculation based on image dimensions and stride. rg 'max_edge_length' sleap_nn/inference/paf_grouping.pyLength of output: 1042
There was a problem hiding this comment.
Actionable comments posted: 0
Review Details
Configuration used: CodeRabbit UI
Review profile: CHILL
Files selected for processing (4)
- docs/config.md (4 hunks)
- sleap_nn/architectures/model.py (2 hunks)
- sleap_nn/inference/predictors.py (11 hunks)
- sleap_nn/training/model_trainer.py (4 hunks)
Files skipped from review as they are similar to previous changes (2)
- sleap_nn/architectures/model.py
- sleap_nn/training/model_trainer.py
Additional Context Used
LanguageTool (42)
docs/config.md (42)
Near line 6: Loose punctuation mark.
Context: ... four main sections: - 1.data_config: Creating a data pipeline. - 2. `model_...
Near line 8: Loose punctuation mark.
Context: ...ng a data pipeline. - 2.model_config: Initialise the sleap-nn backbone and he...
Near line 10: Loose punctuation mark.
Context: ... and head models. - 3.trainer_config: Hyperparameters required to train the m...
Near line 12: Loose punctuation mark.
Context: ...with Lightning. - 4.inference_config: Inference related configs. Note:...
Near line 16: Loose punctuation mark.
Context: ... forval_data_loader. -data_config: -provider: (str) Provider class...
Near line 19: Loose punctuation mark.
Context: ...psPipeline" or "BottomUp". -train: -labels_path: (str) Path to ...
Near line 21: Possible missing article found.
Context: ...he image has 3 channels (RGB image). If input has only one channel when this ...
Near line 22: Possible missing article found.
Context: ... is set toTrue, then the images from single-channel is replicated along the...
Near line 23: Possible missing article found.
Context: ...s replicated along the channel axis. If input has three channels and this is ...
Near line 24: Possible missing article found.
Context: ... to False, then we convert the image to grayscale (single-channel) image. ...
Near line 31: Loose punctuation mark.
Context: ...e same factor. -preprocessing: -anchor_ind: (int) Index...
Near line 32: Possible missing comma found.
Context: ...can significantly improve topdown model accuracy as they benefit from a consistent geome...
Near line 34: ‘with respect to’ might be wordy. Consider a shorter alternative.
Context: ...re easier to learn but are less precise with respect to the peak coordinate. This spread is in ...
Near line 36: Loose punctuation mark.
Context: ...n. -augmentation_config: -random crop`: (Dict[...
Near line 64: Loose punctuation mark.
Context: ... totrainstructure) -model_config: -init_weight: (str) model weigh...
Near line 67: Loose punctuation mark.
Context: ...win_B_Weights"]. -backbone_config: -backbone_type: (str) Backbo...
Near line 78: This phrase might be redundant. Consider either removing or replacing the adjective ‘additional’.
Context: ... -middle_block: (bool) If True, add an additional block at the end of the encoder. default: Tru...
Near line 80: Possible missing comma found.
Context: ... for upsampling. Interpolation is faster but transposed convolutions may...
Near line 103: Possible missing comma found.
Context: ... for upsampling. Interpolation is faster but transposed convolutions may...
Near line 108: Loose punctuation mark.
Context: .... Default: "tiny". -arch: Dictionary of embed dimension, depths a...
Near line 120: Possible missing comma found.
Context: ... for upsampling. Interpolation is faster but transposed convolutions may...
Near line 123: Possible missing comma found.
Context: ...List[dict]) List of heads in the model. For eg, BottomUp model has both 'MultiInsta...
Near line 128: Possible missing comma found.
Context: ...can significantly improve topdown model accuracy as they benefit from a consistent geome...
Near line 129: ‘with respect to’ might be wordy. Consider a shorter alternative.
Context: ...re easier to learn but are less precise with respect to the peak coordinate. This spread is in ...
Near line 139: Possible missing article found.
Context: ...he batch size. If False and the size of dataset is not divisible by the batch size, the...
Near line 143: Possible typo: you repeated a word
Context: ...ease note that the monitors are checked every every_n_epochs epochs. if save_top_k >= 2 and...
Near line 143: Possible typo: you repeated a word
Context: ... the monitors are checked every every_n_epochs epochs. if save_top_k >= 2 and the callback is...
Near line 173: Possible missing comma found.
Context: ...onitored has stopped decreasing; in max mode it will be reduced when the quantity mo...
Near line 181: Loose punctuation mark.
Context: ...ely. Default: 0. -inference_config: -device: (str) Device on which t...
Near line 183: Loose punctuation mark.
Context: ... "ideep", "hip", "msnpu"). -data: -path: (str) Path to.slp...
Near line 190: Possible missing article found.
Context: ...he image has 3 channels (RGB image). If input has only one channel when this ...
Near line 191: Possible missing article found.
Context: ... is set toTrue, then the images from single-channel is replicated along the...
Near line 192: Possible missing article found.
Context: ...s replicated along the channel axis. If input has three channels and this is ...
Near line 193: Possible missing article found.
Context: ... to False, then we convert the image to grayscale (single-channel) image. ...
Near line 203: Loose punctuation mark.
Context: ... the default. -preprocessing: -anchor_ind: (int) Inde...
Near line 204: Possible missing comma found.
Context: ...can significantly improve topdown model accuracy as they benefit from a consistent geome...
Near line 208: Loose punctuation mark.
Context: ...the input image. -peak_threshold:floatbetween 0 and 1. Minimum confid...
Near line 209: Loose punctuation mark.
Context: ... be ignored. -integral_refinement: IfNone, returns the grid-aligned pea...
Near line 210: Loose punctuation mark.
Context: ... regression. -integral_patch_size: Size of patches to crop around each rou...
Near line 211: Loose punctuation mark.
Context: ... integer scalar. -return_confmaps: IfTrue, predicted confidence maps wi...
Near line 212: Loose punctuation mark.
Context: ... values and points. -return_pafs: IfTrue, predicted part affinity fiel...
Near line 213: Loose punctuation mark.
Context: ...es and points. -return_paf_graph: IfTrue, the part affinity field grap...
Markdownlint (177)
docs/config.md (177)
17: Expected: 2; Actual: 4
Unordered list indentation
18: Expected: 2; Actual: 4
Unordered list indentation
19: Expected: 2; Actual: 4
Unordered list indentation
20: Expected: 4; Actual: 8
Unordered list indentation
21: Expected: 4; Actual: 8
Unordered list indentation
26: Expected: 4; Actual: 8
Unordered list indentation
28: Expected: 4; Actual: 8
Unordered list indentation
30: Expected: 4; Actual: 8
Unordered list indentation
31: Expected: 4; Actual: 8
Unordered list indentation
32: Expected: 6; Actual: 12
Unordered list indentation
33: Expected: 6; Actual: 12
Unordered list indentation
34: Expected: 6; Actual: 12
Unordered list indentation
35: Expected: 6; Actual: 12
Unordered list indentation
36: Expected: 6; Actual: 12
Unordered list indentation
37: Expected: 8; Actual: 16
Unordered list indentation
38: Expected: 8; Actual: 16
Unordered list indentation
39: Expected: 8; Actual: 16
Unordered list indentation
40: Expected: 10; Actual: 20
Unordered list indentation
41: Expected: 12; Actual: 24
Unordered list indentation
42: Expected: 12; Actual: 24
Unordered list indentation
43: Expected: 12; Actual: 24
Unordered list indentation
44: Expected: 12; Actual: 24
Unordered list indentation
45: Expected: 12; Actual: 24
Unordered list indentation
46: Expected: 12; Actual: 24
Unordered list indentation
47: Expected: 12; Actual: 24
Unordered list indentation
48: Expected: 12; Actual: 24
Unordered list indentation
49: Expected: 12; Actual: 24
Unordered list indentation
50: Expected: 10; Actual: 20
Unordered list indentation
51: Expected: 12; Actual: 24
Unordered list indentation
52: Expected: 12; Actual: 24
Unordered list indentation
54: Expected: 12; Actual: 24
Unordered list indentation
55: Expected: 12; Actual: 24
Unordered list indentation
56: Expected: 12; Actual: 24
Unordered list indentation
57: Expected: 12; Actual: 24
Unordered list indentation
58: Expected: 12; Actual: 24
Unordered list indentation
59: Expected: 12; Actual: 24
Unordered list indentation
60: Expected: 12; Actual: 24
Unordered list indentation
61: Expected: 12; Actual: 24
Unordered list indentation
62: Expected: 2; Actual: 4
Unordered list indentation
65: Expected: 2; Actual: 4
Unordered list indentation
66: Expected: 2; Actual: 4
Unordered list indentation
67: Expected: 2; Actual: 4
Unordered list indentation
68: Expected: 4; Actual: 8
Unordered list indentation
69: Expected: 4; Actual: 8
Unordered list indentation
70: Expected: 6; Actual: 12
Unordered list indentation
71: Expected: 6; Actual: 12
Unordered list indentation
72: Expected: 6; Actual: 12
Unordered list indentation
73: Expected: 6; Actual: 12
Unordered list indentation
74: Expected: 6; Actual: 12
Unordered list indentation
76: Expected: 6; Actual: 12
Unordered list indentation
78: Expected: 6; Actual: 12
Unordered list indentation
79: Expected: 6; Actual: 12
Unordered list indentation
83: Expected: 6; Actual: 12
Unordered list indentation
84: Expected: 6; Actual: 12
Unordered list indentation
88: Expected: 6; Actual: 12
Unordered list indentation
89: Expected: 6; Actual: 12
Unordered list indentation
90: Expected: 4; Actual: 8
Unordered list indentation
91: Expected: 6; Actual: 12
Unordered list indentation
92: Expected: 8; Actual: 16
Unordered list indentation
93: Expected: 8; Actual: 16
Unordered list indentation
94: Expected: 6; Actual: 12
Unordered list indentation
95: Expected: 6; Actual: 12
Unordered list indentation
96: Expected: 6; Actual: 12
Unordered list indentation
97: Expected: 6; Actual: 12
Unordered list indentation
98: Expected: 6; Actual: 12
Unordered list indentation
99: Expected: 6; Actual: 12
Unordered list indentation
100: Expected: 6; Actual: 12
Unordered list indentation
101: Expected: 6; Actual: 12
Unordered list indentation
102: Expected: 6; Actual: 12
Unordered list indentation
106: Expected: 4; Actual: 8
Unordered list indentation
107: Expected: 6; Actual: 12
Unordered list indentation
108: Expected: 6; Actual: 12
Unordered list indentation
111: Expected: 6; Actual: 12
Unordered list indentation
112: Expected: 6; Actual: 12
Unordered list indentation
113: Expected: 6; Actual: 12
Unordered list indentation
114: Expected: 6; Actual: 12
Unordered list indentation
115: Expected: 6; Actual: 12
Unordered list indentation
116: Expected: 6; Actual: 12
Unordered list indentation
117: Expected: 6; Actual: 12
Unordered list indentation
118: Expected: 6; Actual: 12
Unordered list indentation
119: Expected: 6; Actual: 12
Unordered list indentation
123: Expected: 2; Actual: 4
Unordered list indentation
134: Expected: 2; Actual: 4
Unordered list indentation
135: Expected: 4; Actual: 8
Unordered list indentation
136: Expected: 4; Actual: 8
Unordered list indentation
137: Expected: 4; Actual: 8
Unordered list indentation
138: Expected: 4; Actual: 8
Unordered list indentation
139: Expected: 4; Actual: 8
Unordered list indentation
140: Expected: 4; Actual: 8
Unordered list indentation
141: Expected: 2; Actual: 4
Unordered list indentation
142: Expected: 2; Actual: 4
Unordered list indentation
143: Expected: 4; Actual: 8
Unordered list indentation
144: Expected: 4; Actual: 8
Unordered list indentation
145: Expected: 4; Actual: 8
Unordered list indentation
146: Expected: 4; Actual: 8
Unordered list indentation
147: Expected: 4; Actual: 8
Unordered list indentation
148: Expected: 2; Actual: 4
Unordered list indentation
149: Expected: 4; Actual: 8
Unordered list indentation
150: Expected: 4; Actual: 8
Unordered list indentation
151: Expected: 4; Actual: 8
Unordered list indentation
152: Expected: 2; Actual: 4
Unordered list indentation
153: Expected: 2; Actual: 4
Unordered list indentation
154: Expected: 2; Actual: 4
Unordered list indentation
155: Expected: 2; Actual: 4
Unordered list indentation
156: Expected: 2; Actual: 4
Unordered list indentation
157: Expected: 2; Actual: 4
Unordered list indentation
158: Expected: 2; Actual: 4
Unordered list indentation
159: Expected: 2; Actual: 4
Unordered list indentation
160: Expected: 2; Actual: 4
Unordered list indentation
161: Expected: 2; Actual: 4
Unordered list indentation
162: Expected: 4; Actual: 8
Unordered list indentation
163: Expected: 4; Actual: 8
Unordered list indentation
164: Expected: 4; Actual: 8
Unordered list indentation
165: Expected: 4; Actual: 8
Unordered list indentation
166: Expected: 4; Actual: 8
Unordered list indentation
167: Expected: 4; Actual: 8
Unordered list indentation
168: Expected: 2; Actual: 4
Unordered list indentation
169: Expected: 2; Actual: 4
Unordered list indentation
170: Expected: 4; Actual: 8
Unordered list indentation
171: Expected: 4; Actual: 8
Unordered list indentation
172: Expected: 2; Actual: 4
Unordered list indentation
173: Expected: 4; Actual: 8
Unordered list indentation
174: Expected: 4; Actual: 8
Unordered list indentation
175: Expected: 4; Actual: 8
Unordered list indentation
176: Expected: 4; Actual: 8
Unordered list indentation
177: Expected: 4; Actual: 8
Unordered list indentation
178: Expected: 4; Actual: 8
Unordered list indentation
179: Expected: 4; Actual: 8
Unordered list indentation
182: Expected: 2; Actual: 4
Unordered list indentation
183: Expected: 2; Actual: 4
Unordered list indentation
184: Expected: 4; Actual: 8
Unordered list indentation
185: Expected: 4; Actual: 8
Unordered list indentation
187: Expected: 4; Actual: 8
Unordered list indentation
189: Expected: 4; Actual: 8
Unordered list indentation
190: Expected: 4; Actual: 8
Unordered list indentation
195: Expected: 4; Actual: 8
Unordered list indentation
196: Expected: 4; Actual: 8
Unordered list indentation
197: Expected: 4; Actual: 8
Unordered list indentation
198: Expected: 6; Actual: 12
Unordered list indentation
199: Expected: 6; Actual: 12
Unordered list indentation
200: Expected: 6; Actual: 12
Unordered list indentation
201: Expected: 6; Actual: 12
Unordered list indentation
203: Expected: 4; Actual: 8
Unordered list indentation
204: Expected: 6; Actual: 12
Unordered list indentation
205: Expected: 6; Actual: 12
Unordered list indentation
206: Expected: 6; Actual: 12
Unordered list indentation
207: Expected: 6; Actual: 12
Unordered list indentation
208: Expected: 2; Actual: 4
Unordered list indentation
209: Expected: 2; Actual: 4
Unordered list indentation
210: Expected: 2; Actual: 4
Unordered list indentation
211: Expected: 2; Actual: 4
Unordered list indentation
212: Expected: 2; Actual: 4
Unordered list indentation
213: Expected: 2; Actual: 4
Unordered list indentation
16: Expected: 0 or 2; Actual: 1
Trailing spaces
33: Expected: 0 or 2; Actual: 1
Trailing spaces
35: Expected: 0 or 2; Actual: 1
Trailing spaces
39: Expected: 0 or 2; Actual: 1
Trailing spaces
40: Expected: 0 or 2; Actual: 1
Trailing spaces
64: Expected: 0 or 2; Actual: 1
Trailing spaces
94: Expected: 0 or 2; Actual: 1
Trailing spaces
107: Expected: 0 or 2; Actual: 1
Trailing spaces
127: Expected: 0 or 2; Actual: 1
Trailing spaces
133: Expected: 0 or 2; Actual: 1
Trailing spaces
159: Expected: 0 or 2; Actual: 1
Trailing spaces
182: Expected: 0 or 2; Actual: 1
Trailing spaces
194: Expected: 0 or 2; Actual: 1
Trailing spaces
203: Expected: 0 or 2; Actual: 1
Trailing spaces
207: Expected: 0 or 2; Actual: 1
Trailing spaces
211: Expected: 0 or 2; Actual: 1
Trailing spaces
212: Expected: 0 or 2; Actual: 1
Trailing spaces
54: null
Spaces inside emphasis markers
54: null
Spaces inside emphasis markers
54: null
Spaces inside emphasis markers
54: null
Spaces inside emphasis markers
175: null
Spaces inside emphasis markers
175: null
Spaces inside emphasis markers
213: null
Files should end with a single newline character
Additional comments not posted (10)
docs/config.md (4)
18-18: Add "BottomUp" to the list of pipeline options.This addition aligns with the PR's objective to introduce a BottomUp model pipeline, ensuring the documentation is consistent with the new feature.
35-35: Add "pafs_gen" configuration for the BottomUp model.This configuration is crucial for defining the generation of Part Affinity Fields (PAFs), which are essential for the BottomUp model's functionality.
123-131: Update "head_configs" to include new head types for the BottomUp model.The inclusion of 'PartAffinityFieldsHead' and 'MultiInstanceConfmapsHead' is necessary for the BottomUp model to function correctly, as these components handle different aspects of the model's output.
207-213: Add new inference configuration options.The addition of
pafs_output_stride,return_confmaps,return_pafs, andreturn_paf_graphprovides necessary controls for the inference process, particularly for handling output from the BottomUp model.sleap_nn/inference/predictors.py (6)
33-33: Added import forBottomUpModel.This import is necessary for the new
BottomUpPredictorclass introduced in this PR.
37-37: Added import forBottomUpInferenceModel.This import is necessary for the new
BottomUpPredictorclass to function correctly.
98-104: Introduced handling forPartAffinityFieldsHeadin model type detection.This change is crucial for supporting the new
BottomUpPredictorwhich relies onPartAffinityFieldsHead.
155-161: Handling nested tensors during prediction output processing.This addition ensures that nested tensors are correctly converted to numpy arrays, which is important for compatibility with other parts of the system that expect numpy arrays.
Also applies to: 207-213
441-456: Added aKeyFilterto the data pipeline to ensure only necessary keys are passed through.This is a good practice to ensure that only relevant data is processed, which can improve performance and reduce memory usage.
846-992: Introduction ofBottomUpPredictorclass.This class is well-structured and aligns with the objectives of the PR to add bottom-up model pipeline capabilities. It handles initialization, preprocessing, and prediction for a trained BottomUp SLEAP-NN model.
There was a problem hiding this comment.
Actionable comments posted: 1
Outside diff range and nitpick comments (3)
tests/architectures/test_encoder_decoder.py (2)
Line range hint
10-10: Add a docstring to describe the purpose and functionality oftest_simple_conv_block.+ """Test the SimpleConvBlock for correct layer types and sequence."""
Line range hint
33-33: Add a docstring to describe the purpose and functionality oftest_simple_upsampling_block.+ """Test the SimpleUpsamplingBlock for correct upsampling behavior and output shapes."""tests/architectures/test_swint.py (1)
Line range hint
1-1: Add a module docstring to provide an overview of the file's purpose and scope.+ """Tests for Swin Transformer architecture."""
Review Details
Configuration used: CodeRabbit UI
Review profile: CHILL
Files selected for processing (8)
- sleap_nn/inference/paf_grouping.py (4 hunks)
- sleap_nn/inference/predictors.py (14 hunks)
- tests/architectures/test_encoder_decoder.py (1 hunks)
- tests/architectures/test_swint.py (1 hunks)
- tests/data/test_normalization.py (2 hunks)
- tests/inference/test_predictors.py (1 hunks)
- tests/inference/test_topdown.py (1 hunks)
- tests/training/test_model_trainer.py (3 hunks)
Files skipped from review as they are similar to previous changes (1)
- sleap_nn/inference/paf_grouping.py
Additional Context Used
Ruff (9)
tests/architectures/test_encoder_decoder.py (3)
1-1: Missing docstring in public module
10-10: Missing docstring in public function
33-33: Missing docstring in public function
tests/architectures/test_swint.py (1)
1-1: Missing docstring in public module
tests/data/test_normalization.py (1)
1-1: Missing docstring in public module
tests/inference/test_predictors.py (1)
1-1: Missing docstring in public module
tests/inference/test_topdown.py (1)
1-1: Missing docstring in public module
tests/training/test_model_trainer.py (2)
71-71: Missing docstring in public function
271-271: Missing docstring in public function
Additional comments not posted (14)
tests/data/test_normalization.py (1)
26-30: The grayscale conversion test is correctly implemented.tests/architectures/test_encoder_decoder.py (1)
123-123: The addition oftranspose_convs_batch_norm_before_activationenhances model configurability.tests/architectures/test_swint.py (1)
196-240: The tests for custom architecture configurations are comprehensive and correctly validate the expected outputs.tests/inference/test_topdown.py (1)
26-66: Theinitialize_modelfunction is correctly implemented to set up the model and data pipeline for inference testing.tests/training/test_model_trainer.py (1)
Line range hint
226-430: Thetest_bottomup_modelfunction is correctly implemented to test the BottomUp model training with comprehensive checks on output shapes and configurations.tests/inference/test_predictors.py (3)
8-32: LGTM! The test is well-structured and covers various assertions to validate the functionality of theTopDownPredictor.
196-236: LGTM! The test is well-structured and covers various assertions to validate the functionality of theSingleInstancePredictor.
311-435: LGTM! The test is well-structured and covers various assertions to validate the functionality of theBottomUpPredictor.sleap_nn/inference/predictors.py (6)
33-33: TheBottomUpModelclass has been added to the import list.This addition aligns with the PR's objective to enhance the framework with bottom-up model capabilities.
37-37: Introduction of theBottomUpInferenceModelclass.This is a crucial addition for supporting the new bottom-up inference capabilities as described in the PR summary.
98-104: Addition of theBottomUpPredictorto handle specific model types.This change is essential for supporting the new bottom-up model types and is consistent with the overall goal of the PR.
153-159: Handling of nested tensors during prediction output processing.This ensures that the output tensors are correctly handled, especially when they are nested, which is important for the robustness of the inference process.
195-211: Enhanced preprocessing and tensor handling in theVideoReaderdata provider.These changes are crucial for ensuring that video data is correctly preprocessed and that tensors are properly managed during batch processing.
837-1176: Introduction of theBottomUpPredictorclass.This new class is central to the implementation of the bottom-up model prediction capabilities. It handles initialization, preprocessing, and prediction, aligning perfectly with the PR's objectives to enhance the framework's capabilities.
There was a problem hiding this comment.
Actionable comments posted: 0
Review Details
Configuration used: CodeRabbit UI
Review profile: CHILL
Files selected for processing (2)
- sleap_nn/inference/predictors.py (17 hunks)
- tests/inference/test_predictors.py (1 hunks)
Additional Context Used
Ruff (1)
tests/inference/test_predictors.py (1)
1-1: Missing docstring in public module
Additional comments not posted (18)
tests/inference/test_predictors.py (5)
1-1: Add a module-level docstring to describe the purpose and scope of the tests in this file.
8-12: The test functiontest_topdown_predictoris well-documented and the logic appears sound. Ensure that the test covers all relevant cases for theTopDownPredictor.
237-237: Add a docstring totest_single_instance_predictorto explain the purpose of the test and its parameters.
353-353: Add a docstring totest_bottomup_predictorto explain the purpose of the test and its parameters.
8-8: Add a docstring totest_topdown_predictorto explain the purpose of the test and its parameters.sleap_nn/inference/predictors.py (13)
883-943: The initialization method for theBottomUpPredictoris well-implemented, correctly setting up the model and PAF scorer based on the provided configurations.
951-986: The methodfrom_trained_modelscorrectly handles model loading and configuration, ensuring that the model is set to the appropriate device.
988-1068: Themake_pipelinemethod is robust, handling different data providers and correctly setting up the data pipeline.
1070-1180: The method_make_labeled_frames_from_generatoreffectively converts generator output into SLEAP-specific labeled frames, handling the data structure conversion appropriately.
945-948: Thedata_configproperty method correctly retrieves the data configuration from the overall configuration.
Line range hint
283-304: The initialization method for theTopDownPredictoris well-implemented, setting up the model layers based on the provided configurations.
Line range hint
283-304: The methodfrom_trained_modelsforTopDownPredictorcorrectly handles model loading and configuration, ensuring that the model is set to the appropriate device.
Line range hint
283-304: Themake_pipelinemethod forTopDownPredictoris robust, handling different data providers and correctly setting up the data pipeline.
Line range hint
283-304: The method_make_labeled_frames_from_generatorforTopDownPredictoreffectively converts generator output into SLEAP-specific labeled frames, handling the data structure conversion appropriately.
Line range hint
283-304: The initialization method for theSingleInstancePredictoris well-implemented, setting up the model based on the provided configurations.
Line range hint
283-304: The methodfrom_trained_modelsforSingleInstancePredictorcorrectly handles model loading and configuration, ensuring that the model is set to the appropriate device.
Line range hint
283-304: Themake_pipelinemethod forSingleInstancePredictoris robust, handling different data providers and correctly setting up the data pipeline.
Line range hint
283-304: The method_make_labeled_frames_from_generatorforSingleInstancePredictoreffectively converts generator output into SLEAP-specific labeled frames, handling the data structure conversion appropriately.
There was a problem hiding this comment.
Actionable comments posted: 0
Review Details
Configuration used: CodeRabbit UI
Review profile: CHILL
Files selected for processing (3)
- sleap_nn/architectures/encoder_decoder.py (1 hunks)
- tests/architectures/test_encoder_decoder.py (1 hunks)
- tests/architectures/test_unet.py (1 hunks)
Additional Context Used
Ruff (5)
tests/architectures/test_encoder_decoder.py (3)
1-1: Missing docstring in public module
10-10: Missing docstring in public function
33-33: Missing docstring in public function
tests/architectures/test_unet.py (2)
1-1: Missing docstring in public module
10-10: Missing docstring in public function
Additional comments not posted (3)
tests/architectures/test_encoder_decoder.py (1)
108-136: Added new configuration parameters toSimpleUpsamplingBlockfor more flexible upsampling strategies. Ensure that these changes are reflected in the corresponding documentation and that all dependent code is updated accordingly.tests/architectures/test_unet.py (1)
174-199: Introducedblock_contractioninEncoderconfiguration to potentially optimize model performance. Verify that this new configuration does not adversely affect the accuracy or performance of the model in real-world scenarios.sleap_nn/architectures/encoder_decoder.py (1)
240-255: Refined theblock_contractionlogic in theEncoderclass to allow for a configurable reduction in the number of filters. This enhancement could lead to more efficient models, especially in resource-constrained environments.
There was a problem hiding this comment.
Actionable comments posted: 0
Outside diff range and nitpick comments (1)
tests/architectures/test_unet.py (1)
Line range hint
10-10: Please add a docstring totest_unet_referenceto explain its purpose and the specific aspects of UNet it tests.
Review Details
Configuration used: CodeRabbit UI
Review profile: CHILL
Files selected for processing (2)
- sleap_nn/inference/paf_grouping.py (4 hunks)
- tests/architectures/test_unet.py (1 hunks)
Files skipped from review as they are similar to previous changes (1)
- sleap_nn/inference/paf_grouping.py
Additional Context Used
Ruff (2)
tests/architectures/test_unet.py (2)
1-1: Missing docstring in public module
10-10: Missing docstring in public function
Additional comments not posted (1)
tests/architectures/test_unet.py (1)
174-199: The addition of the block contraction test is well-integrated and correctly checks the expected outputs and features. Good job ensuring comprehensive testing coverage for this configuration.
There was a problem hiding this comment.
Actionable comments posted: 8
Review Details
Configuration used: CodeRabbit UI
Review profile: CHILL
Files selected for processing (10)
- sleap_nn/data/augmentation.py (2 hunks)
- sleap_nn/data/confidence_maps.py (11 hunks)
- sleap_nn/data/instance_centroids.py (1 hunks)
- sleap_nn/data/instance_cropping.py (3 hunks)
- sleap_nn/data/normalization.py (1 hunks)
- sleap_nn/data/pipelines.py (2 hunks)
- sleap_nn/data/providers.py (2 hunks)
- sleap_nn/inference/bottomup.py (1 hunks)
- sleap_nn/inference/predictors.py (1 hunks)
- tests/inference/test_predictors.py (1 hunks)
Files skipped from review due to trivial changes (2)
- sleap_nn/data/instance_centroids.py
- sleap_nn/data/normalization.py
Files skipped from review as they are similar to previous changes (3)
- sleap_nn/data/confidence_maps.py
- sleap_nn/data/instance_cropping.py
- sleap_nn/data/providers.py
Additional Context Used
Ruff (1)
tests/inference/test_predictors.py (1)
1-1: Missing docstring in public module
Additional comments not posted (11)
sleap_nn/inference/bottomup.py (2)
10-78: The implementation ofBottomUpInferenceModelis comprehensive and well-documented.
103-160: Theforwardmethod is well-structured and handles multiple outputs effectively.sleap_nn/data/augmentation.py (1)
Line range hint
294-310: Update in__iter__method correctly handles reshaping of instances.sleap_nn/data/pipelines.py (1)
293-381: TheBottomUpPipelineclass is well-implemented and aligns with the PR objectives.tests/inference/test_predictors.py (4)
1-1: Add a module-level docstring to describe the purpose and scope of the tests in this file.
8-8: Add a docstring totest_topdown_predictorto explain the purpose of the test and its parameters.
237-237: Add a docstring totest_single_instance_predictorto explain the purpose of the test and its parameters.
353-353: Add a docstring totest_bottomup_predictorto explain the purpose of the test and its parameters.sleap_nn/inference/predictors.py (3)
44-79: The base classPredictoris well-defined with clear documentation and attributes. Good use of type hints for clarity.
162-170: The method_convert_tensors_to_numpyefficiently handles tensor conversion. Ensure thatis_nestedattribute is properly defined in the tensor class.
286-371: TheTopDownPredictorclass is well-documented and the methods are clearly defined. However, ensure that thecentroid_configandconfmap_configare always synchronized when used.Verification successful
The search results indicate that both
centroid_configandconfmap_configare used multiple times within thesleap_nn/inference/predictors.pyfile. Importantly, there is a conditional checkif self.centroid_config and self.confmap_config:which suggests that the synchronization between these two configurations is being handled.
centroid_configis used in various contexts, including model initialization and configuration settings.confmap_configis similarly used for model initialization and configuration settings.The presence of the conditional check
if self.centroid_config and self.confmap_config:ensures that both configurations are considered together when necessary.Based on this evidence, it appears that the synchronization concern is addressed in the code.
Scripts Executed
The following scripts were executed for the analysis:
Script:
#!/bin/bash # Description: Verify synchronization of `centroid_config` and `confmap_config`. # Test: Search for usage of these configs to ensure they are used together. Expect: Consistent usage. rg --type python 'centroid_config' 'sleap_nn/inference/predictors.py' rg --type python 'confmap_config' 'sleap_nn/inference/predictors.py'Length of output: 196
Script:
#!/bin/bash # Description: Verify synchronization of `centroid_config` and `confmap_config`. # Test: Search for usage of these configs to ensure they are used together. Expect: Consistent usage. rg 'centroid_config' sleap_nn/inference/predictors.py rg 'confmap_config' sleap_nn/inference/predictors.pyLength of output: 5182
Co-authored-by: coderabbitai[bot] <136622811+coderabbitai[bot]@users.noreply.github.com>
There was a problem hiding this comment.
Actionable comments posted: 5
Review Details
Configuration used: CodeRabbit UI
Review profile: CHILL
Files selected for processing (1)
- sleap_nn/inference/predictors.py (1 hunks)
Additional comments not posted (1)
sleap_nn/inference/predictors.py (1)
44-79: The classPredictoris well-documented and serves as a base for other predictors. Good use of Python'sattrslibrary for concise attribute definitions.
There was a problem hiding this comment.
Actionable comments posted: 0
Review Details
Configuration used: CodeRabbit UI
Review profile: CHILL
Files selected for processing (2)
- docs/config.md (4 hunks)
- sleap_nn/inference/predictors.py (1 hunks)
Additional Context Used
LanguageTool (47)
docs/config.md (47)
Near line 6: Loose punctuation mark.
Context: ... four main sections: - 1.data_config: Creating a data pipeline. - 2. `model_...
Rule ID: UNLIKELY_OPENING_PUNCTUATION
Near line 8: Loose punctuation mark.
Context: ...ng a data pipeline. - 2.model_config: Initialise the sleap-nn backbone and he...
Rule ID: UNLIKELY_OPENING_PUNCTUATION
Near line 10: Loose punctuation mark.
Context: ... and head models. - 3.trainer_config: Hyperparameters required to train the m...
Rule ID: UNLIKELY_OPENING_PUNCTUATION
Near line 12: Loose punctuation mark.
Context: ...with Lightning. - 4.inference_config: Inference related configs. Note:...
Rule ID: UNLIKELY_OPENING_PUNCTUATION
Near line 16: Loose punctuation mark.
Context: ... forval_data_loader. -data_config: -provider: (str) Provider class...
Rule ID: UNLIKELY_OPENING_PUNCTUATION
Near line 19: Loose punctuation mark.
Context: ...psPipeline" or "BottomUp". -train: -labels_path: (str) Path to ...
Rule ID: UNLIKELY_OPENING_PUNCTUATION
Near line 21: Possible missing article found.
Context: ...he image has 3 channels (RGB image). If input has only one channel when this ...
Rule ID: AI_HYDRA_LEO_MISSING_THE
Near line 22: Possible missing article found.
Context: ... is set toTrue, then the images from single-channel is replicated along the...
Rule ID: AI_HYDRA_LEO_MISSING_THE
Near line 23: Possible missing article found.
Context: ...s replicated along the channel axis. If input has three channels and this is ...
Rule ID: AI_HYDRA_LEO_MISSING_THE
Near line 24: Possible missing article found.
Context: ... to False, then we convert the image to grayscale (single-channel) image. ...
Rule ID: AI_HYDRA_LEO_MISSING_A
Near line 31: Loose punctuation mark.
Context: ...e same factor. -preprocessing: -anchor_ind: (int) Index...
Rule ID: UNLIKELY_OPENING_PUNCTUATION
Near line 32: Possible missing comma found.
Context: ...can significantly improve topdown model accuracy as they benefit from a consistent geome...
Rule ID: AI_HYDRA_LEO_MISSING_COMMA
Near line 34: Possible missing comma found.
Context: ...space. Larger values are easier to learn but are less precise with respect to the pe...
Rule ID: AI_HYDRA_LEO_MISSING_COMMA
Near line 34: ‘with respect to’ might be wordy. Consider a shorter alternative.
Context: ...re easier to learn but are less precise with respect to the peak coordinate. This spread is in ...
Rule ID: EN_WORDINESS_PREMIUM_WITH_RESPECT_TO
Near line 36: Loose punctuation mark.
Context: ...n. -augmentation_config: -random crop`: (Dict[...
Rule ID: UNLIKELY_OPENING_PUNCTUATION
Near line 64: Loose punctuation mark.
Context: ... totrainstructure) -model_config: -init_weight: (str) model weigh...
Rule ID: UNLIKELY_OPENING_PUNCTUATION
Near line 67: Loose punctuation mark.
Context: ...win_B_Weights"]. -backbone_config: -backbone_type: (str) Backbo...
Rule ID: UNLIKELY_OPENING_PUNCTUATION
Near line 78: This phrase might be redundant. Consider either removing or replacing the adjective ‘additional’.
Context: ... -middle_block: (bool) If True, add an additional block at the end of the encoder. default: Tru...
Rule ID: ADD_AN_ADDITIONAL
Near line 80: Possible missing comma found.
Context: ... for upsampling. Interpolation is faster but transposed convolutions may...
Rule ID: AI_HYDRA_LEO_MISSING_COMMA
Near line 103: Possible missing comma found.
Context: ... for upsampling. Interpolation is faster but transposed convolutions may...
Rule ID: AI_HYDRA_LEO_MISSING_COMMA
Near line 108: Loose punctuation mark.
Context: .... Default: "tiny". -arch: Dictionary of embed dimension, depths a...
Rule ID: UNLIKELY_OPENING_PUNCTUATION
Near line 120: Possible missing comma found.
Context: ... for upsampling. Interpolation is faster but transposed convolutions may...
Rule ID: AI_HYDRA_LEO_MISSING_COMMA
Near line 123: Possible missing comma found.
Context: ...List[dict]) List of heads in the model. For eg, BottomUp model has both 'MultiInsta...
Rule ID: AI_HYDRA_LEO_MISSING_COMMA
Near line 128: Possible missing comma found.
Context: ...can significantly improve topdown model accuracy as they benefit from a consistent geome...
Rule ID: AI_HYDRA_LEO_MISSING_COMMA
Near line 129: Possible missing comma found.
Context: ...space. Larger values are easier to learn but are less precise with respect to the pe...
Rule ID: AI_HYDRA_LEO_MISSING_COMMA
Near line 129: ‘with respect to’ might be wordy. Consider a shorter alternative.
Context: ...re easier to learn but are less precise with respect to the peak coordinate. This spread is in ...
Rule ID: EN_WORDINESS_PREMIUM_WITH_RESPECT_TO
Near line 139: Possible missing article found.
Context: ...he batch size. If False and the size of dataset is not divisible by the batch size, the...
Rule ID: AI_HYDRA_LEO_MISSING_THE
Near line 143: Possible typo: you repeated a word
Context: ...ease note that the monitors are checked every every_n_epochs epochs. if save_top_k >= 2 and...
Rule ID: ENGLISH_WORD_REPEAT_RULE
Near line 143: Possible typo: you repeated a word
Context: ... the monitors are checked every every_n_epochs epochs. if save_top_k >= 2 and the callback is...
Rule ID: ENGLISH_WORD_REPEAT_RULE
Near line 146: Possible missing comma found.
Context: ... -monitor: (str) Quantity to monitor for e.g., "val_loss". When None, this saves...
Rule ID: AI_HYDRA_LEO_MISSING_COMMA
Near line 173: Possible missing comma found.
Context: ...onitored has stopped decreasing; in max mode it will be reduced when the quantity mo...
Rule ID: AI_HYDRA_LEO_MISSING_COMMA
Near line 177: Possible missing comma found.
Context: ...tience`: (int) Number of epochs with no improvement after which learning rate will be reduc...
Rule ID: AI_HYDRA_LEO_MISSING_COMMA
Near line 179: Possible missing comma found.
Context: ...arning rate of all param groups or each group respectively. Default: 0. - `inferen...
Rule ID: AI_HYDRA_LEO_MISSING_COMMA
Near line 181: Loose punctuation mark.
Context: ...ely. Default: 0. -inference_config: -device: (str) Device on which t...
Rule ID: UNLIKELY_OPENING_PUNCTUATION
Near line 183: Loose punctuation mark.
Context: ... "ideep", "hip", "msnpu"). -data: -path: (str) Path to.slp...
Rule ID: UNLIKELY_OPENING_PUNCTUATION
Near line 190: Possible missing article found.
Context: ...he image has 3 channels (RGB image). If input has only one channel when this ...
Rule ID: AI_HYDRA_LEO_MISSING_THE
Near line 191: Possible missing article found.
Context: ... is set toTrue, then the images from single-channel is replicated along the...
Rule ID: AI_HYDRA_LEO_MISSING_THE
Near line 192: Possible missing article found.
Context: ...s replicated along the channel axis. If input has three channels and this is ...
Rule ID: AI_HYDRA_LEO_MISSING_THE
Near line 193: Possible missing article found.
Context: ... to False, then we convert the image to grayscale (single-channel) image. ...
Rule ID: AI_HYDRA_LEO_MISSING_A
Near line 203: Loose punctuation mark.
Context: ... the default. -preprocessing: -anchor_ind: (int) Inde...
Rule ID: UNLIKELY_OPENING_PUNCTUATION
Near line 204: Possible missing comma found.
Context: ...can significantly improve topdown model accuracy as they benefit from a consistent geome...
Rule ID: AI_HYDRA_LEO_MISSING_COMMA
Near line 208: Loose punctuation mark.
Context: ...the input image. -peak_threshold:floatbetween 0 and 1. Minimum confid...
Rule ID: UNLIKELY_OPENING_PUNCTUATION
Near line 209: Loose punctuation mark.
Context: ... be ignored. -integral_refinement: IfNone, returns the grid-aligned pea...
Rule ID: UNLIKELY_OPENING_PUNCTUATION
Near line 210: Loose punctuation mark.
Context: ... regression. -integral_patch_size: Size of patches to crop around each rou...
Rule ID: UNLIKELY_OPENING_PUNCTUATION
Near line 211: Loose punctuation mark.
Context: ... integer scalar. -return_confmaps: IfTrue, predicted confidence maps wi...
Rule ID: UNLIKELY_OPENING_PUNCTUATION
Near line 212: Loose punctuation mark.
Context: ... values and points. -return_pafs: IfTrue, predicted part affinity fiel...
Rule ID: UNLIKELY_OPENING_PUNCTUATION
Near line 213: Loose punctuation mark.
Context: ...es and points. -return_paf_graph: IfTrue, the part affinity field grap...
Rule ID: UNLIKELY_OPENING_PUNCTUATION
Markdownlint (185)
docs/config.md (185)
17: Expected: 2; Actual: 4
Unordered list indentation
18: Expected: 2; Actual: 4
Unordered list indentation
19: Expected: 2; Actual: 4
Unordered list indentation
20: Expected: 4; Actual: 8
Unordered list indentation
21: Expected: 4; Actual: 8
Unordered list indentation
26: Expected: 4; Actual: 8
Unordered list indentation
28: Expected: 4; Actual: 8
Unordered list indentation
30: Expected: 4; Actual: 8
Unordered list indentation
31: Expected: 4; Actual: 8
Unordered list indentation
32: Expected: 6; Actual: 12
Unordered list indentation
33: Expected: 6; Actual: 12
Unordered list indentation
34: Expected: 6; Actual: 12
Unordered list indentation
35: Expected: 6; Actual: 12
Unordered list indentation
36: Expected: 6; Actual: 12
Unordered list indentation
37: Expected: 8; Actual: 16
Unordered list indentation
38: Expected: 8; Actual: 16
Unordered list indentation
39: Expected: 8; Actual: 16
Unordered list indentation
40: Expected: 10; Actual: 20
Unordered list indentation
41: Expected: 12; Actual: 24
Unordered list indentation
42: Expected: 12; Actual: 24
Unordered list indentation
43: Expected: 12; Actual: 24
Unordered list indentation
44: Expected: 12; Actual: 24
Unordered list indentation
45: Expected: 12; Actual: 24
Unordered list indentation
46: Expected: 12; Actual: 24
Unordered list indentation
47: Expected: 12; Actual: 24
Unordered list indentation
48: Expected: 12; Actual: 24
Unordered list indentation
49: Expected: 12; Actual: 24
Unordered list indentation
50: Expected: 10; Actual: 20
Unordered list indentation
51: Expected: 12; Actual: 24
Unordered list indentation
52: Expected: 12; Actual: 24
Unordered list indentation
54: Expected: 12; Actual: 24
Unordered list indentation
55: Expected: 12; Actual: 24
Unordered list indentation
56: Expected: 12; Actual: 24
Unordered list indentation
57: Expected: 12; Actual: 24
Unordered list indentation
58: Expected: 12; Actual: 24
Unordered list indentation
59: Expected: 12; Actual: 24
Unordered list indentation
60: Expected: 12; Actual: 24
Unordered list indentation
61: Expected: 12; Actual: 24
Unordered list indentation
62: Expected: 2; Actual: 4
Unordered list indentation
65: Expected: 2; Actual: 4
Unordered list indentation
66: Expected: 2; Actual: 4
Unordered list indentation
67: Expected: 2; Actual: 4
Unordered list indentation
68: Expected: 4; Actual: 8
Unordered list indentation
69: Expected: 4; Actual: 8
Unordered list indentation
70: Expected: 6; Actual: 12
Unordered list indentation
71: Expected: 6; Actual: 12
Unordered list indentation
72: Expected: 6; Actual: 12
Unordered list indentation
73: Expected: 6; Actual: 12
Unordered list indentation
74: Expected: 6; Actual: 12
Unordered list indentation
76: Expected: 6; Actual: 12
Unordered list indentation
78: Expected: 6; Actual: 12
Unordered list indentation
79: Expected: 6; Actual: 12
Unordered list indentation
83: Expected: 6; Actual: 12
Unordered list indentation
84: Expected: 6; Actual: 12
Unordered list indentation
88: Expected: 6; Actual: 12
Unordered list indentation
89: Expected: 6; Actual: 12
Unordered list indentation
90: Expected: 4; Actual: 8
Unordered list indentation
91: Expected: 6; Actual: 12
Unordered list indentation
92: Expected: 8; Actual: 16
Unordered list indentation
93: Expected: 8; Actual: 16
Unordered list indentation
94: Expected: 6; Actual: 12
Unordered list indentation
95: Expected: 6; Actual: 12
Unordered list indentation
96: Expected: 6; Actual: 12
Unordered list indentation
97: Expected: 6; Actual: 12
Unordered list indentation
98: Expected: 6; Actual: 12
Unordered list indentation
99: Expected: 6; Actual: 12
Unordered list indentation
100: Expected: 6; Actual: 12
Unordered list indentation
101: Expected: 6; Actual: 12
Unordered list indentation
102: Expected: 6; Actual: 12
Unordered list indentation
106: Expected: 4; Actual: 8
Unordered list indentation
107: Expected: 6; Actual: 12
Unordered list indentation
108: Expected: 6; Actual: 12
Unordered list indentation
111: Expected: 6; Actual: 12
Unordered list indentation
112: Expected: 6; Actual: 12
Unordered list indentation
113: Expected: 6; Actual: 12
Unordered list indentation
114: Expected: 6; Actual: 12
Unordered list indentation
115: Expected: 6; Actual: 12
Unordered list indentation
116: Expected: 6; Actual: 12
Unordered list indentation
117: Expected: 6; Actual: 12
Unordered list indentation
118: Expected: 6; Actual: 12
Unordered list indentation
119: Expected: 6; Actual: 12
Unordered list indentation
123: Expected: 2; Actual: 4
Unordered list indentation
124: Expected: 4; Actual: 8
Unordered list indentation
125: Expected: 4; Actual: 8
Unordered list indentation
126: Expected: 6; Actual: 12
Unordered list indentation
127: Expected: 6; Actual: 12
Unordered list indentation
128: Expected: 6; Actual: 12
Unordered list indentation
129: Expected: 6; Actual: 12
Unordered list indentation
130: Expected: 6; Actual: 12
Unordered list indentation
131: Expected: 6; Actual: 12
Unordered list indentation
134: Expected: 2; Actual: 4
Unordered list indentation
135: Expected: 4; Actual: 8
Unordered list indentation
136: Expected: 4; Actual: 8
Unordered list indentation
137: Expected: 4; Actual: 8
Unordered list indentation
138: Expected: 4; Actual: 8
Unordered list indentation
139: Expected: 4; Actual: 8
Unordered list indentation
140: Expected: 4; Actual: 8
Unordered list indentation
141: Expected: 2; Actual: 4
Unordered list indentation
142: Expected: 2; Actual: 4
Unordered list indentation
143: Expected: 4; Actual: 8
Unordered list indentation
144: Expected: 4; Actual: 8
Unordered list indentation
145: Expected: 4; Actual: 8
Unordered list indentation
146: Expected: 4; Actual: 8
Unordered list indentation
147: Expected: 4; Actual: 8
Unordered list indentation
148: Expected: 2; Actual: 4
Unordered list indentation
149: Expected: 4; Actual: 8
Unordered list indentation
150: Expected: 4; Actual: 8
Unordered list indentation
151: Expected: 4; Actual: 8
Unordered list indentation
152: Expected: 2; Actual: 4
Unordered list indentation
153: Expected: 2; Actual: 4
Unordered list indentation
154: Expected: 2; Actual: 4
Unordered list indentation
155: Expected: 2; Actual: 4
Unordered list indentation
156: Expected: 2; Actual: 4
Unordered list indentation
157: Expected: 2; Actual: 4
Unordered list indentation
158: Expected: 2; Actual: 4
Unordered list indentation
159: Expected: 2; Actual: 4
Unordered list indentation
160: Expected: 2; Actual: 4
Unordered list indentation
161: Expected: 2; Actual: 4
Unordered list indentation
162: Expected: 4; Actual: 8
Unordered list indentation
163: Expected: 4; Actual: 8
Unordered list indentation
164: Expected: 4; Actual: 8
Unordered list indentation
165: Expected: 4; Actual: 8
Unordered list indentation
166: Expected: 4; Actual: 8
Unordered list indentation
167: Expected: 4; Actual: 8
Unordered list indentation
168: Expected: 2; Actual: 4
Unordered list indentation
169: Expected: 2; Actual: 4
Unordered list indentation
170: Expected: 4; Actual: 8
Unordered list indentation
171: Expected: 4; Actual: 8
Unordered list indentation
172: Expected: 2; Actual: 4
Unordered list indentation
173: Expected: 4; Actual: 8
Unordered list indentation
174: Expected: 4; Actual: 8
Unordered list indentation
175: Expected: 4; Actual: 8
Unordered list indentation
176: Expected: 4; Actual: 8
Unordered list indentation
177: Expected: 4; Actual: 8
Unordered list indentation
178: Expected: 4; Actual: 8
Unordered list indentation
179: Expected: 4; Actual: 8
Unordered list indentation
182: Expected: 2; Actual: 4
Unordered list indentation
183: Expected: 2; Actual: 4
Unordered list indentation
184: Expected: 4; Actual: 8
Unordered list indentation
185: Expected: 4; Actual: 8
Unordered list indentation
187: Expected: 4; Actual: 8
Unordered list indentation
189: Expected: 4; Actual: 8
Unordered list indentation
190: Expected: 4; Actual: 8
Unordered list indentation
195: Expected: 4; Actual: 8
Unordered list indentation
196: Expected: 4; Actual: 8
Unordered list indentation
197: Expected: 4; Actual: 8
Unordered list indentation
198: Expected: 6; Actual: 12
Unordered list indentation
199: Expected: 6; Actual: 12
Unordered list indentation
200: Expected: 6; Actual: 12
Unordered list indentation
201: Expected: 6; Actual: 12
Unordered list indentation
203: Expected: 4; Actual: 8
Unordered list indentation
204: Expected: 6; Actual: 12
Unordered list indentation
205: Expected: 6; Actual: 12
Unordered list indentation
206: Expected: 6; Actual: 12
Unordered list indentation
207: Expected: 6; Actual: 12
Unordered list indentation
208: Expected: 2; Actual: 4
Unordered list indentation
209: Expected: 2; Actual: 4
Unordered list indentation
210: Expected: 2; Actual: 4
Unordered list indentation
211: Expected: 2; Actual: 4
Unordered list indentation
212: Expected: 2; Actual: 4
Unordered list indentation
213: Expected: 2; Actual: 4
Unordered list indentation
16: Expected: 0 or 2; Actual: 1
Trailing spaces
33: Expected: 0 or 2; Actual: 1
Trailing spaces
35: Expected: 0 or 2; Actual: 1
Trailing spaces
39: Expected: 0 or 2; Actual: 1
Trailing spaces
40: Expected: 0 or 2; Actual: 1
Trailing spaces
64: Expected: 0 or 2; Actual: 1
Trailing spaces
94: Expected: 0 or 2; Actual: 1
Trailing spaces
107: Expected: 0 or 2; Actual: 1
Trailing spaces
127: Expected: 0 or 2; Actual: 1
Trailing spaces
133: Expected: 0 or 2; Actual: 1
Trailing spaces
159: Expected: 0 or 2; Actual: 1
Trailing spaces
182: Expected: 0 or 2; Actual: 1
Trailing spaces
194: Expected: 0 or 2; Actual: 1
Trailing spaces
203: Expected: 0 or 2; Actual: 1
Trailing spaces
207: Expected: 0 or 2; Actual: 1
Trailing spaces
211: Expected: 0 or 2; Actual: 1
Trailing spaces
212: Expected: 0 or 2; Actual: 1
Trailing spaces
54: null
Spaces inside emphasis markers
54: null
Spaces inside emphasis markers
54: null
Spaces inside emphasis markers
54: null
Spaces inside emphasis markers
175: null
Spaces inside emphasis markers
175: null
Spaces inside emphasis markers
213: null
Files should end with a single newline character
Additional comments not posted (12)
docs/config.md (8)
18-18: Add "BottomUp" to the pipeline options.This addition aligns with the PR's objective to introduce a comprehensive pipeline for BottomUp models.
35-35: Addition ofpafs_genconfiguration for the BottomUp model.This configuration is crucial for defining the generation of Part Affinity Fields, which are essential for the BottomUp approach.
123-123: Update tohead_configsto include new configurations for the BottomUp model.The inclusion of 'PartAffinityFieldsHead' and 'MultiInstanceConfmapsHead' is necessary for the new model's functionality.
126-126: Specify body parts for model heads.This configuration allows for flexible specification of body parts, which is important for customizing model outputs.
127-127: Define edges for 'PartAffinityFieldsHead'.This is a critical addition for the BottomUp model as it defines the connections between body parts.
128-128: Anchor point configuration for 'CenteredInstanceConfmapsHead'.Properly setting the anchor point can significantly enhance model accuracy by ensuring consistent geometry.
207-207: Addpafs_output_strideconfiguration.This setting is important for defining the resolution of the output part affinity fields, which impacts both performance and accuracy.
211-213: Return additional outputs for inference.These options (
return_confmaps,return_pafs,return_paf_graph) provide flexibility in the output, allowing users to choose what additional data they need post-inference.sleap_nn/inference/predictors.py (4)
883-1218: Addition ofBottomUpPredictorclass.The newly added
BottomUpPredictorclass is well-structured and aligns with the objectives of the PR to enhance the BottomUp model pipeline. The methods within the class are well-documented and the attributes are clearly defined. Good use of type annotations for clarity.
655-881: Addition ofSingleInstancePredictorclass.The
SingleInstancePredictorclass has been successfully added and is consistent with the PR's aim to improve inference capabilities. The class methods are appropriately abstracted and the documentation is clear, aiding in maintainability and understanding.
286-652: Refactor ofTopDownPredictorclass.The refactoring of the
TopDownPredictorclass appears comprehensive and well-executed. The changes are in line with the PR's description of enhancing the inference modules. The class now handles multiple configurations and model types more robustly.
44-284: General enhancements inPredictorbase class.The modifications in the
Predictorbase class, including the addition of new attributes and methods, enhance its functionality and flexibility. These changes support a wider range of model types and configurations, which is crucial for the scalability of the inference process.
There was a problem hiding this comment.
Actionable comments posted: 11
Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Files selected for processing (1)
- sleap_nn/inference/predictors.py (1 hunks)
Additional context used
Ruff
sleap_nn/inference/predictors.py
82-82:
typing.Textis deprecated, usestr(UP019)Replace with
str
244-244: Within an
exceptclause, raise exceptions withraise ... from errorraise ... from Noneto distinguish them from errors in exception handling (B904)
386-386:
typing.Textis deprecated, usestr(UP019)Replace with
str
387-387:
typing.Textis deprecated, usestr(UP019)Replace with
str
575-575: Use
key in dictinstead ofkey in dict.keys()(SIM118)Remove
.keys()
616-616: Loop control variable
org_sizenot used within loop body (B007)Rename unused
org_sizeto_org_size
697-697:
typing.Textis deprecated, usestr(UP019)Replace with
str
813-813: Use
key in dictinstead ofkey in dict.keys()(SIM118)Remove
.keys()
851-851: Loop control variable
org_sizenot used within loop body (B007)Rename unused
org_sizeto_org_size
951-952: Use
key in dictinstead ofkey in dict.keys()(SIM118)Remove
.keys()
957-957: Use
key in dictinstead ofkey in dict.keys()(SIM118)Remove
.keys()
962-962: Use
key in dictinstead ofkey in dict.keys()(SIM118)Remove
.keys()
967-967: Use
key in dictinstead ofkey in dict.keys()(SIM118)Remove
.keys()
972-972: Use
key in dictinstead ofkey in dict.keys()(SIM118)Remove
.keys()
1004-1004:
typing.Textis deprecated, usestr(UP019)Replace with
str
1125-1125: Use
key in dictinstead ofkey in dict.keys()(SIM118)Remove
.keys()
This PR adds the data processing, model training and inference pipeline for BottomUp models where all the nodes across all instances are predicted at one-shot and the nodes belonging to a certain instance are combine suing the
inference.paf_grouping.PAFScorermodule.The inference modules are refactored, addressing #48.
Summary by CodeRabbit
New Features
TopDownPredictor,SingleInstancePredictor, andBottomUpPredictor.Documentation