File size: 7,912 Bytes
ea8c728 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 | import matplotlib
import pandas as pd
import pytest
matplotlib.use("Agg")
from shinka.plots import (
plot_generation_runtime_timeline,
plot_normalized_occupancy_over_time,
)
from shinka.plots.plot_throughput import (
_compute_occupancy_series,
_prepare_pool_runtime_data,
)
def _runtime_df() -> pd.DataFrame:
return pd.DataFrame(
[
{
"id": "job-a-main",
"source_job_id": "job-a",
"is_island_copy": False,
"correct": True,
"combined_score": 0.9,
"timestamp": 10,
"generation": 1,
"patch_name": "patch-a",
"model_name": "model-a",
"timeline_lane_mode": "pool_slots",
"pipeline_started_at": 0,
"sampling_started_at": 0,
"sampling_finished_at": 2,
"evaluation_started_at": 2,
"evaluation_finished_at": 6,
"postprocess_started_at": 6,
"postprocess_finished_at": 8,
"sampling_worker_id": 1,
"evaluation_worker_id": 1,
"postprocess_worker_id": 1,
"sampling_worker_capacity": 2,
"evaluation_worker_capacity": 2,
"postprocess_worker_capacity": 1,
},
{
"id": "job-a-copy",
"source_job_id": "job-a",
"is_island_copy": True,
"correct": False,
"combined_score": 0.1,
"timestamp": 11,
"generation": 1,
"patch_name": "patch-a-copy",
"model_name": "model-a",
"timeline_lane_mode": "pool_slots",
"pipeline_started_at": 0,
"sampling_started_at": 0,
"sampling_finished_at": 2,
"evaluation_started_at": 2,
"evaluation_finished_at": 6,
"postprocess_started_at": 6,
"postprocess_finished_at": 8,
"sampling_worker_id": 1,
"evaluation_worker_id": 1,
"postprocess_worker_id": 1,
"sampling_worker_capacity": 2,
"evaluation_worker_capacity": 2,
"postprocess_worker_capacity": 1,
},
{
"id": "job-b",
"source_job_id": "job-b",
"is_island_copy": False,
"correct": True,
"combined_score": 0.8,
"timestamp": 20,
"generation": 2,
"patch_name": "patch-b",
"model_name": "model-b",
"timeline_lane_mode": "pool_slots",
"pipeline_started_at": 1,
"sampling_started_at": 1,
"sampling_finished_at": 3,
"evaluation_started_at": 4,
"evaluation_finished_at": 8,
"postprocess_started_at": 8,
"postprocess_finished_at": 10,
"sampling_worker_id": 2,
"evaluation_worker_id": 2,
"postprocess_worker_id": 1,
"sampling_worker_capacity": 2,
"evaluation_worker_capacity": 2,
"postprocess_worker_capacity": 1,
},
{
"id": "job-missing",
"source_job_id": "job-missing",
"is_island_copy": False,
"correct": True,
"combined_score": 0.7,
"timestamp": 30,
"generation": 3,
"patch_name": "patch-missing",
"model_name": "model-c",
"timeline_lane_mode": "pool_slots",
"pipeline_started_at": 2,
"sampling_started_at": 2,
"sampling_finished_at": 4,
"evaluation_started_at": 5,
"evaluation_finished_at": 9,
"postprocess_started_at": 9,
"postprocess_finished_at": None,
"sampling_worker_id": 1,
"evaluation_worker_id": 1,
"postprocess_worker_id": 1,
"sampling_worker_capacity": 2,
"evaluation_worker_capacity": 2,
"postprocess_worker_capacity": 1,
},
]
)
def test_prepare_pool_runtime_data_dedupes_rows_and_computes_capacities():
prepared = _prepare_pool_runtime_data(_runtime_df())
assert prepared is not None
assert prepared.capacities == {"sampling": 2, "evaluation": 2, "postprocess": 1}
assert prepared.lane_labels == [
"Sampling W1",
"Evaluation W1",
"Postprocess W1",
"Sampling W2",
"Evaluation W2",
]
assert list(prepared.rows["id"]) == ["job-a-main", "job-b"]
assert prepared.peaks == {"sampling": 2, "evaluation": 2, "postprocess": 1}
def test_prepare_pool_runtime_data_handles_missing_optional_columns():
runtime_df = _runtime_df().drop(
columns=["source_job_id", "is_island_copy", "patch_name", "model_name"]
)
prepared = _prepare_pool_runtime_data(runtime_df)
assert prepared is not None
assert list(prepared.rows["id"]) == ["job-a-main", "job-a-copy", "job-b"]
assert list(prepared.rows["source_job_id"]) == ["job-a-main", "job-a-copy", "job-b"]
assert list(prepared.rows["patch_name"]) == ["unnamed", "unnamed", "unnamed"]
assert list(prepared.rows["model_name"]) == ["N/A", "N/A", "N/A"]
assert prepared.capacities == {"sampling": 2, "evaluation": 2, "postprocess": 1}
def test_compute_occupancy_series_matches_expected_utilization_stats():
prepared = _prepare_pool_runtime_data(_runtime_df())
assert prepared is not None
series = _compute_occupancy_series(
prepared.rows,
start_key="evaluation_started_at",
end_key="evaluation_finished_at",
capacity=prepared.capacities["evaluation"],
)
assert series is not None
assert series.total_duration == pytest.approx(6.0)
assert series.avg_occupied == pytest.approx(8.0 / 6.0)
assert series.utilization_pct == pytest.approx((8.0 / 12.0) * 100.0)
assert series.full_occupancy_pct == pytest.approx((2.0 / 6.0) * 100.0)
assert series.idle_pct == pytest.approx(0.0)
def test_plot_generation_runtime_timeline_uses_deduped_pool_rows():
fig, ax = plot_generation_runtime_timeline(_runtime_df(), title="Runtime Timeline")
assert fig is not None
assert ax is not None
assert [tick.get_text() for tick in ax.get_yticklabels()] == [
"Sampling W1",
"Evaluation W1",
"Postprocess W1",
"Sampling W2",
"Evaluation W2",
]
assert len(ax.patches) == 6
assert {text.get_text() for text in ax.get_legend().get_texts()} == {
"Sampling",
"Evaluation",
"Postprocess",
}
assert ax.get_legend()._ncols == 3
assert ax.get_legend()._loc == 9
assert {text.get_fontsize() for text in ax.get_legend().get_texts()} == {10.0}
assert ax.get_legend().get_bbox_to_anchor()._bbox.y0 < 0
def test_plot_normalized_occupancy_over_time_adds_reference_line():
fig, ax = plot_normalized_occupancy_over_time(
_runtime_df(), title="Normalized Occupancy"
)
assert fig is not None
assert ax is not None
labels = [line.get_label() for line in ax.lines]
assert labels == [
"Sampling Occupancy",
"Evaluation Occupancy",
"Postprocess Occupancy",
"100% Capacity",
]
assert ax.get_ylim()[1] >= 100
assert list(ax.lines[-1].get_ydata()) == [100, 100]
assert ax.get_legend()._ncols == 2
assert ax.get_legend()._loc == 9
assert {text.get_fontsize() for text in ax.get_legend().get_texts()} == {10.0}
assert ax.get_legend().get_bbox_to_anchor()._bbox.y0 < 0
|