Skip to content

Instantly share code, notes, and snippets.

@mmore500

mmore500/reconstruction-quality-experiment.ipynb

Last active February 1, 2024 17:24
Show Gist options
  • Save mmore500/a2e88e7c239935c362ec59c6b5a3f7b5 to your computer and use it in GitHub Desktop.
Save mmore500/a2e88e7c239935c362ec59c6b5a3f7b5 to your computer and use it in GitHub Desktop.
Download ZIP
reconstruction-quality-experiment.ipynb
Display the source blob
Display the rendered blob
Raw
reconstruction-quality-experiment.ipynb
{
"nbformat": 4,
"nbformat_minor": 0,
"metadata": {
"colab": {
"provenance": [],
"include_colab_link": true
},
"kernelspec": {
"name": "python3",
"display_name": "Python 3"
},
"language_info": {
"name": "python"
}
},
"cells": [
{
"cell_type": "markdown",
"metadata": {
"id": "view-in-github",
"colab_type": "text"
},
"source": [
"<a href=\"https://colab.research.google.com/gist/mmore500/a2e88e7c239935c362ec59c6b5a3f7b5/reconstruction-quality-experiment.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
]
},
{
"cell_type": "markdown",
"source": [
"## Procedure:\n",
"\n",
"For each experimental replicate per treatment,\n",
"- Navigate to <https://colab.research.google.com/gist/mmore500/a2e88e7c239935c362ec59c6b5a3f7b5> to open a fresh copy of the experiment notebook. **Open a fresh notebook copy for each treatment.**\n",
"- Click on filename on the top left of the Colab page(`a2e88e7c239935c362ec59c6b5a3f7b5`) and rename according to template\n",
" - `evo=island{num_islands}-niche{num_niches}-ngen{num_generations}-popsize{population_size}-tournsize{tournament_size}+instrument={\"steady\"|\"tilted\"}-{\"old\"|\"new\"}-bits{annotation_size_bits}-diff{differentia_width}+replicate={replicate}+ext=.ipynb`.\n",
" - For example, `evo=island1-niche1-ngen10000-popsize1024-tournsize2+instrument=steady-old-bits64-diff1+replicate=0+ext=.ipynb`.\n",
"- Configure variables in \"Configure Experment\" section.\n",
"- On the top menu, click `Runtime > Restart sesson and run all` if available, otherwise `Runtime > Run all`.\n",
"- Wait for final cell's execution to complete.\n",
"- Record configured variables and results from \"Evaluate Reconstruction\" section in [results spreadsheet](https://docs.google.com/spreadsheets/d/1ZhS4NDTDyBiwmwtWrZO5L06MGB3lhmp2-5ZzClhEwPU/edit?usp=sharing).\n",
"- On the top menu, click `File > Download > Download .ipynb`.\n",
"- Upload ipynb file to treatment directory at <https://osf.io/n4b2g/>, named same as notebook, except excluding `+replicate={replicate}+ext=.ipynb`.\n",
" - Treatment directory should contain notebooks for each replicate of notebook.\n"
],
"metadata": {
"id": "lhNERG3shuFb"
}
},
{
"cell_type": "markdown",
"source": [
"## Set Up Environment"
],
"metadata": {
"id": "pzVsZRrQEgVa"
}
},
{
"cell_type": "code",
"source": [
"!python3 -m pip install \\\n",
" \"alifedata_phyloinformatics_convert==0.15.1\" \\\n",
" \"biopython==1.83\" \\\n",
" \"dendropy==4.6.1\" \\\n",
" \"git+https://github.com/mmore500/[email protected]#egg=hsurf\" \\\n",
" \"hstrat==1.9.1\" \\\n",
" \"matplotlib==3.8.2\" \\\n",
" \"pandas==1.5.3\" \\\n",
" \"tqdist==1.0\" \\\n",
" \"tqdm==4.66.1\" \\\n",
" \"typing_extensions>=4.9.0\" \\\n",
" \"watermark==2.4.3\""
],
"metadata": {
"id": "yGzTyOj4EfuD"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"source": [
"from collections import Counter\n",
"import typing\n",
"\n",
"import alifedata_phyloinformatics_convert as apc\n",
"from Bio import Phylo\n",
"import dendropy as dp\n",
"from hstrat import hstrat\n",
"from hstrat import _auxiliary_lib as hstrat_aux\n",
"from hsurf import hsurf\n",
"from matplotlib import pyplot as plt\n",
"import numpy as np\n",
"import pandas as pd\n",
"import tqdist\n",
"from tqdm.notebook import tqdm"
],
"metadata": {
"id": "cIrNBKme_8Ey"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "markdown",
"source": [
"## Configure Experiment"
],
"metadata": {
"id": "H0US4pJ6Cpz4"
}
},
{
"cell_type": "markdown",
"source": [
"Configure instrumentation. **Edit me**"
],
"metadata": {
"id": "hgSx_K65JBqH"
}
},
{
"cell_type": "code",
"source": [
"# TODO Uncomment one...\n",
"# annotation_size_bits = 64\n",
"# annotation_size_bits = 256\n",
"# annotation_size_bits = 1024\n",
"assert annotation_size_bits.bit_count() == 1, \"must be power of 2 (1, 2, 4, 8, etc.)\"\n",
"\n",
"# TODO Uncomment one...\n",
"# differentia_width_bits = 1\n",
"# differentia_width_bits = 8\n",
"assert differentia_width_bits.bit_count() == 1, \"must be power of 2 (1, 2, 4, 8, etc.)\"\n",
"\n",
"# TODO Uncomment one...\n",
"# stratum_retention_algo = hstrat.depth_proportional_resolution_tapered_algo # old impl/steady behavior\n",
"# stratum_retention_algo = hstrat.recency_proportional_resolution_curbed_algo # old impl/tilted behavior\n",
"# stratum_retention_algo = hsurf.stratum_retention_interop_hybrid_algo # new impl/hybrid behavior\n",
"# stratum_retention_algo = hsurf.stratum_retention_interop_steady_algo # new impl/steady behavior\n",
"# stratum_retention_algo = hsurf.stratum_retention_interop_tilted_sticky_algo # new impl/tilted behavior"
],
"metadata": {
"id": "mWGLaLxyJEf8"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "markdown",
"source": [
"Configure evolutionary scale. **Edit me**"
],
"metadata": {
"id": "VhnKN1ktJHYp"
}
},
{
"cell_type": "code",
"source": [
"# TODO Uncomment one...\n",
"# population_size = 1024 # default condition\n",
"# population_size = 65536 # alternate condition\n",
"assert population_size.bit_count() == 1, \"must be power of 2 (1, 2, 4, 8, etc.)\"\n",
"\n",
"# TODO Uncomment one...\n",
"# num_generations = 10000 # default condition\n",
"# num_generations = 100000 # alternate condition\n"
],
"metadata": {
"id": "xQ21R33LCplC"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "markdown",
"source": [
"Configure evolutionary conditions. **Edit me**"
],
"metadata": {
"id": "vtNMCtNcJPLw"
}
},
{
"cell_type": "code",
"source": [
"# TODO Uncomment one...\n",
"# num_islands=1 # default condition\n",
"# num_islands=64 # alternate condition\n",
"assert num_islands.bit_count() == 1, \"must be power of 2 (1, 2, 4, 8, etc.)\"\n",
"\n",
"# TODO Uncomment one...\n",
"# num_niches=1 # default condition\n",
"# num_niches=8 # alternate condition\n",
"assert num_niches.bit_count() == 1, \"must be power of 2 (1, 2, 4, 8, etc.)\"\n",
"\n",
"# TODO Uncomment one...\n",
"# tournament_size=2 # default condition\n",
"# tournament_size=1 # alternate condition\n",
"# tournament_size=8 # alternate condition\n"
],
"metadata": {
"id": "v5nvGaCJJNaI"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "markdown",
"source": [
"Configure experimental replicate. **Edit me**"
],
"metadata": {
"id": "cIrbQgJ_cLhf"
}
},
{
"cell_type": "code",
"source": [
"replicate = # TODO set to a number, 0 through 19"
],
"metadata": {
"id": "g8vC-VBLcKt1"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "markdown",
"source": [
"Set up random number generator. (Do not edit.)"
],
"metadata": {
"id": "pqY6ejDEJgZA"
}
},
{
"cell_type": "code",
"source": [
"seed = hash(\n",
" (\n",
" replicate,\n",
" population_size,\n",
" num_generations,\n",
" num_islands,\n",
" num_niches,\n",
" tournament_size,\n",
" )\n",
") % 2 ** 32\n",
"\n",
"seed"
],
"metadata": {
"id": "naUfe9mNCcsV"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"source": [
"from hstrat._auxiliary_lib import seed_random\n",
"\n",
"seed_random(seed)\n"
],
"metadata": {
"id": "em_mdO9-ENsr"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "markdown",
"source": [
"Parametrize instrumentation. (Do not edit.)"
],
"metadata": {
"id": "nzxql5RHXlrU"
}
},
{
"cell_type": "code",
"source": [
"annotation_capacity_strata = annotation_size_bits // differentia_width_bits\n",
"assert annotation_capacity_strata.bit_count() == 1, \"must be power of 2 (1, 2, 4, 8, etc.)\"\n",
"print(f\"{annotation_capacity_strata=}\")\n",
"\n",
"parametrized_policy = stratum_retention_algo.Policy(\n",
" parameterizer=hstrat.PropertyAtMostParameterizer(\n",
" target_value=annotation_capacity_strata,\n",
" policy_evaluator=hstrat.NumStrataRetainedUpperBoundEvaluator(\n",
" at_num_strata_deposited=num_generations,\n",
" ),\n",
" param_lower_bound=2,\n",
" param_upper_bound=1024,\n",
" ),\n",
")\n",
"\n",
"print(f\"{parametrized_policy=}\")\n",
"print(f\"num strata retained upper bound {parametrized_policy.CalcNumStrataRetainedUpperBound(num_generations)}\")\n"
],
"metadata": {
"id": "tUkf_biLXsrm"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "markdown",
"source": [
"## Setup"
],
"metadata": {
"id": "rGLUHFxxARm4"
}
},
{
"cell_type": "markdown",
"source": [
"Helper functions."
],
"metadata": {
"id": "PA-DX12UHqqN"
}
},
{
"cell_type": "code",
"source": [
"def calc_tqdist_distance(\n",
" x: pd.DataFrame,\n",
" y: pd.DataFrame,\n",
" progress_wrap: typing.Callable = lambda x: x,\n",
" ) -> float:\n",
" \"\"\"Calculate dissimilarity between two trees. Used to measure how accurate\n",
" tree reconstructions are.\"\"\"\n",
" tree_a = apc.RosettaTree(x).as_dendropy\n",
" tree_b = apc.RosettaTree(y).as_dendropy\n",
"\n",
" # must suppress root unifurcations or tqdist barfs\n",
" # see https://github.com/uym2/tripVote/issues/15\n",
" tree_a.unassign_taxa(exclude_leaves=True)\n",
" tree_a.suppress_unifurcations()\n",
" tree_b.unassign_taxa(exclude_leaves=True)\n",
" tree_b.suppress_unifurcations()\n",
"\n",
" tree_a_taxon_labels = [\n",
" leaf.taxon.label for leaf in progress_wrap(tree_a.leaf_node_iter())\n",
" ]\n",
" tree_b_taxon_labels = [\n",
" leaf.taxon.label for leaf in progress_wrap(tree_b.leaf_node_iter())\n",
" ]\n",
" all(\n",
" progress_wrap(\n",
" zip(tree_a.leaf_node_iter(), tree_b.leaf_node_iter(), strict=True),\n",
" ),\n",
" )\n",
" assert sorted(tree_a_taxon_labels) == sorted(tree_b_taxon_labels)\n",
" assert sorted(tree_a_taxon_labels) == sorted(\n",
" x.loc[hstrat_aux.alifestd_find_leaf_ids(x), \"taxon_label\"],\n",
" )\n",
" assert sorted(tree_a_taxon_labels) == sorted(\n",
" y.loc[hstrat_aux.alifestd_find_leaf_ids(y), \"taxon_label\"],\n",
" )\n",
" for taxon_label in progress_wrap(tree_a_taxon_labels):\n",
" assert taxon_label\n",
" assert taxon_label.strip()\n",
"\n",
" newick_a = tree_a.as_string(schema=\"newick\").strip()\n",
" newick_b = tree_b.as_string(schema=\"newick\").strip()\n",
"\n",
" return {\n",
" \"quartet_distance\": tqdist.quartet_distance(newick_a, newick_b),\n",
" \"quartet_distanc_raw\": tqdist.quartet_distance_raw(newick_a, newick_b),\n",
" \"triplet_distance\": tqdist.triplet_distance(newick_a, newick_b),\n",
" \"triplet_distance_raw\": tqdist.triplet_distance_raw(newick_a, newick_b),\n",
" }\n"
],
"metadata": {
"id": "es5aIEGi9-UG"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "markdown",
"source": [
"## Generate Phylogeny\n",
"\n",
"Use simple evolutionary simulation to generate a phylogenetic history to test reconstruction process on."
],
"metadata": {
"id": "X9hDnZb4BvX1"
}
},
{
"cell_type": "code",
"source": [
"true_phylogeny_df = hstrat.evolve_fitness_trait_population(\n",
" num_islands=num_islands,\n",
" num_niches=num_niches,\n",
" num_generations=num_generations,\n",
" population_size=population_size,\n",
" tournament_size=tournament_size,\n",
" progress_wrap=tqdm,\n",
")"
],
"metadata": {
"id": "JNhgAvB8Bu5p"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"source": [
"true_phylogeny_df[\"taxon_label\"] = true_phylogeny_df[\"loc\"].astype(str)\n",
"true_phylogeny_df = hstrat_aux.alifestd_mark_leaves(true_phylogeny_df, mutate=True)\n",
"true_phylogeny_df.loc[\n",
" ~true_phylogeny_df[\"is_leaf\"], \"taxon_label\"\n",
"] = \"\"\n",
"true_phylogeny_df"
],
"metadata": {
"id": "C4GRtk0Y1tBX"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"source": [
"true_phylogeny_df = hstrat_aux.alifestd_to_working_format(\n",
" hstrat_aux.alifestd_collapse_unifurcations(true_phylogeny_df, mutate=True),\n",
" mutate=True,\n",
").reset_index(drop=True)\n",
"true_phylogeny_df"
],
"metadata": {
"id": "cOdQyGeGAe3P"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "markdown",
"source": [
"## Generate Reconstruction\n",
"\n",
"Generate genome annotations as if tracking phylogeny in distributed environment.\n",
"Then run reconstruction proess to estimate true phylogeny from generated annotations."
],
"metadata": {
"id": "kscgdxYbB19z"
}
},
{
"cell_type": "code",
"source": [
"extant_annotations = hstrat.descend_template_phylogeny_alifestd(\n",
" true_phylogeny_df,\n",
" seed_column=hstrat.HereditaryStratigraphicColumn(parametrized_policy),\n",
" extant_ids=hstrat_aux.alifestd_find_leaf_ids(true_phylogeny_df),\n",
" progress_wrap=tqdm,\n",
")\n",
"\n",
"len(extant_annotations)"
],
"metadata": {
"id": "zIDeuWorB1W8"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"source": [
"reconstructed_phylogeny_df = hstrat.build_tree(\n",
" extant_annotations,\n",
" progress_wrap=tqdm,\n",
" version_pin=hstrat.__version__,\n",
" taxon_labels=true_phylogeny_df.loc[\n",
" hstrat_aux.alifestd_find_leaf_ids(true_phylogeny_df),\n",
" \"taxon_label\",\n",
" ],\n",
")\n",
"reconstructed_phylogeny_df"
],
"metadata": {
"id": "QFCunjmBj9EH"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"source": [
"reconstructed_phylogeny_df = hstrat_aux.alifestd_collapse_unifurcations(reconstructed_phylogeny_df, mutate=True)\n",
"reconstructed_phylogeny_df"
],
"metadata": {
"id": "kDxYkpLvA5SO"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "markdown",
"source": [
"## Evaluate Reconstruction\n",
"\n",
"Reconstruction quality data --- collect into spreadsheet."
],
"metadata": {
"id": "HyyTqjlXB5uu"
}
},
{
"cell_type": "code",
"source": [
"estimation_intervals = [\n",
" hstrat.calc_ranks_since_mrca_bounds_with(\n",
" *np.random.choice(extant_annotations, size=2, replace=False),\n",
" prior=\"arbitrary\",\n",
" )\n",
"\n",
" for __ in tqdm(range(200))\n",
"]\n"
],
"metadata": {
"id": "CN3aqzDS-aT3"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"source": [
"median_abs_uncertainty = np.median([*map(np.ptp, estimation_intervals)])\n",
"mean_abs_uncertainty = np.mean([*map(np.ptp, estimation_intervals)])\n",
"f\"{median_abs_uncertainty=} {mean_abs_uncertainty=}\""
],
"metadata": {
"id": "93JmY11g-eVp"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"source": [
"rel_uncertainties = (\n",
" np.array([*map(np.ptp, estimation_intervals)])\n",
" / (np.array([*map(np.mean, estimation_intervals)]) + 1)\n",
")\n",
"median_rel_uncertainty = np.median(rel_uncertainties)\n",
"mean_rel_uncertainty = np.mean(rel_uncertainties)\n",
"f\"{median_rel_uncertainty=} {mean_rel_uncertainty=}\""
],
"metadata": {
"id": "5kgf112d-4Bq"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"source": [
"num_true_inner_nodes = hstrat_aux.alifestd_count_inner_nodes(true_phylogeny_df)\n",
"num_reconstructed_inner_nodes = hstrat_aux.alifestd_count_inner_nodes(reconstructed_phylogeny_df)\n",
"f\"{num_true_inner_nodes=} {num_reconstructed_inner_nodes=}\""
],
"metadata": {
"id": "staoF6vMDGW3"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"source": [
"num_true_polytomies = hstrat_aux.alifestd_count_polytomies(true_phylogeny_df)\n",
"num_reconstructed_polytomies = hstrat_aux.alifestd_count_polytomies(reconstructed_phylogeny_df)\n",
"f\"{num_true_polytomies=} {num_reconstructed_polytomies=}\""
],
"metadata": {
"id": "5tiqawvDgwqS"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"source": [
"true_polytomic_index = hstrat_aux.alifestd_calc_polytomic_index(true_phylogeny_df)\n",
"reconstructed_polytomic_index = hstrat_aux.alifestd_calc_polytomic_index(reconstructed_phylogeny_df)\n",
"f\"{true_polytomic_index=} {reconstructed_polytomic_index=}\""
],
"metadata": {
"id": "S1XOZAtJAizJ"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"source": [
"distances = calc_tqdist_distance(\n",
" true_phylogeny_df,\n",
" reconstructed_phylogeny_df,\n",
" progress_wrap=tqdm,\n",
")\n",
"f\"{distances=}\""
],
"metadata": {
"id": "ZWJpZlpMB5N0"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"source": [
"sampled_triplet_distance_strict = hstrat_aux.alifestd_estimate_triplet_distance_asexual(\n",
" true_phylogeny_df,\n",
" reconstructed_phylogeny_df,\n",
" taxon_label_key=\"taxon_label\",\n",
" confidence=0.8,\n",
" precision=0.05,\n",
" strict=True,\n",
" progress_wrap=tqdm,\n",
" mutate=True,\n",
")\n",
"f\"{sampled_triplet_distance_strict=}\""
],
"metadata": {
"id": "s7XAQgYwBeTh"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"source": [
"sampled_triplet_distance_lax = hstrat_aux.alifestd_estimate_triplet_distance_asexual(\n",
" true_phylogeny_df,\n",
" reconstructed_phylogeny_df,\n",
" taxon_label_key=\"taxon_label\",\n",
" confidence=0.8,\n",
" precision=0.05,\n",
" strict=False,\n",
" progress_wrap=tqdm,\n",
" mutate=True,\n",
")\n",
"f\"{sampled_triplet_distance_lax=}\""
],
"metadata": {
"id": "2DTLCS4T3yJJ"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "markdown",
"source": [
"## Visualize Phylogeny & Reconstruction\n",
"\n",
"For validating results."
],
"metadata": {
"id": "6J_PwYFnB-Zi"
}
},
{
"cell_type": "markdown",
"source": [
"Topology only (no time)."
],
"metadata": {
"id": "TZSzgxPWzMfE"
}
},
{
"cell_type": "code",
"source": [
"if population_size <= 1024:\n",
" true_phylogeny_tree = apc.alife_dataframe_to_biopython_tree(\n",
" hstrat_aux.alifestd_collapse_unifurcations(true_phylogeny_df),\n",
" setup_branch_lengths=False,\n",
" )\n",
" reconstructed_phylogeny_tree = apc.alife_dataframe_to_biopython_tree(\n",
" hstrat_aux.alifestd_collapse_unifurcations(reconstructed_phylogeny_df),\n",
" setup_branch_lengths=False,\n",
" )\n",
"\n",
" fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(12, 6))\n",
"\n",
" ax1.set_title(\"True Tree\")\n",
" Phylo.draw(true_phylogeny_tree, do_show=False, axes=ax1)\n",
"\n",
" ax2.set_title(\"Reconstructed Tree\")\n",
" Phylo.draw(reconstructed_phylogeny_tree, do_show=False, axes=ax2)\n",
"\n",
" plt.tight_layout()\n",
" plt.show()"
],
"metadata": {
"id": "dhSB7eqTy86F"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "markdown",
"source": [
"Scaled by time."
],
"metadata": {
"id": "zxvNxM9NzVY6"
}
},
{
"cell_type": "code",
"source": [
"if population_size <= 1024:\n",
" true_phylogeny_tree = apc.alife_dataframe_to_biopython_tree(\n",
" hstrat_aux.alifestd_collapse_unifurcations(true_phylogeny_df),\n",
" setup_branch_lengths=True,\n",
" )\n",
" reconstructed_phylogeny_tree = apc.alife_dataframe_to_biopython_tree(\n",
" hstrat_aux.alifestd_collapse_unifurcations(reconstructed_phylogeny_df),\n",
" setup_branch_lengths=True,\n",
" )\n",
"\n",
" fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(12, 6))\n",
"\n",
" ax1.set_title(\"True Tree\")\n",
" ax1.set_xscale(\"log\")\n",
" Phylo.draw(true_phylogeny_tree, do_show=False, axes=ax1)\n",
"\n",
" ax2.set_title(\"Reconstructed Tree\")\n",
" ax2.set_xscale(\"log\")\n",
" Phylo.draw(reconstructed_phylogeny_tree, do_show=False, axes=ax2)\n",
"\n",
" plt.tight_layout()\n",
" plt.show()"
],
"metadata": {
"id": "SKd7tpe5-2e5"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "markdown",
"source": [
"## Reproducibility Information\n",
"\n",
"For future reference if reproducing experiments."
],
"metadata": {
"id": "1vWYtYFg_39e"
}
},
{
"cell_type": "code",
"source": [
"print(\n",
" f\"\"\"# instrumentation\n",
"{annotation_size_bits=}\n",
"{differentia_width_bits=}\n",
"{stratum_retention_algo.PolicySpec.GetAlgoTitle()=}\n",
"\n",
"# evolutionary scale\n",
"{population_size=}\n",
"{num_generations=}\n",
"\n",
"# evolutionary conditions\n",
"{num_islands=}\n",
"{num_niches=}\n",
"{tournament_size=}\n",
"\"\"\"\n",
")"
],
"metadata": {
"id": "q0N8A_8JDRXr"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"source": [
"import datetime\n",
"datetime.datetime.now().isoformat()"
],
"metadata": {
"id": "9rKQfXdBAxxJ"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"source": [
"%load_ext watermark\n",
"%watermark"
],
"metadata": {
"id": "qO0LRJeL_3lB"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"source": [
"!pip freeze"
],
"metadata": {
"id": "UfHdK7OK-afO"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"source": [
"locals()"
],
"metadata": {
"id": "fLVm-ivpDjR6"
},
"execution_count": null,
"outputs": []
}
]
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment