steffen/self-replicating-neural-networks
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

updated journal_soup_basin: all working, only orange lines not showing

Browse Source
This commit is contained in:
ru43zex
2021-06-24 16:48:52 +03:00
parent cf6eec639f
commit f7a0d360b3
1 changed files with 50 additions and 32 deletions
Download Patch File Download Diff File Expand all files Collapse all files

82
journal_soup_basins.py
View File

@@ -1,21 +1,21 @@
import os
from pathlib import Path
import pickle
from torch import mean
from tqdm import tqdm
import random
import copy import copy
from functionalities_test import is_identity_function, test_status, test_for_fixpoints, is_zero_fixpoint, is_divergent, is_secondary_fixpoint import pickle
from network import Net import random
from visualization import plot_3d_self_train, plot_loss, plot_3d_soup from pathlib import Path
import numpy as np import numpy as np
from tabulate import tabulate
from sklearn.metrics import mean_absolute_error as MAE
from sklearn.metrics import mean_squared_error as MSE
import pandas as pd import pandas as pd
import seaborn as sns import seaborn as sns
from matplotlib import pyplot as plt from matplotlib import pyplot as plt
from sklearn.metrics import mean_absolute_error as MAE
from sklearn.metrics import mean_squared_error as MSE
from tabulate import tabulate
from tqdm import tqdm
from functionalities_test import is_identity_function, test_status, is_zero_fixpoint, is_divergent, \
is_secondary_fixpoint
from network import Net
from visualization import plot_loss, plot_3d_soup
def prng(): def prng():
@@ -131,16 +131,16 @@ class SoupSpawnExperiment:
self.populate_environment() self.populate_environment()
self.spawn_and_continue() self.spawn_and_continue()
self.weights_evolution_3d_experiment(self.parents, "only_parents") # self.weights_evolution_3d_experiment(self.parents, "only_parents")
self.weights_evolution_3d_experiment(self.clones, "only_clones") self.weights_evolution_3d_experiment(self.clones, "only_clones")
self.weights_evolution_3d_experiment(self.parents_with_clones, "parents_with_clones") self.weights_evolution_3d_experiment(self.parents_with_clones, "parents_with_clones")
self.weights_evolution_3d_experiment(self.parents_clones_id_functions, "id_f_with_parents") # self.weights_evolution_3d_experiment(self.parents_clones_id_functions, "id_f_with_parents")
# self.visualize_loss() # self.visualize_loss()
self.distance_matrix = distance_matrix(self.parents_clones_id_functions, print_it=False) self.distance_matrix = distance_matrix(self.parents_clones_id_functions, print_it=False)
self.parent_clone_distances = distance_from_parent(self.parents_clones_id_functions, print_it=False) self.parent_clone_distances = distance_from_parent(self.parents_clones_id_functions, print_it=False)
self.save() # self.save()
def populate_environment(self): def populate_environment(self):
loop_population_size = tqdm(range(self.population_size)) loop_population_size = tqdm(range(self.population_size))
@@ -172,7 +172,7 @@ class SoupSpawnExperiment:
def evolve(self, population): def evolve(self, population):
print(f"Clone soup has a population of {len(population)} networks") print(f"Clone soup has a population of {len(population)} networks")
loop_epochs = tqdm(range(self.epochs-1)) loop_epochs = tqdm(range(self.epochs - 1))
for i in loop_epochs: for i in loop_epochs:
loop_epochs.set_description("\nEvolving clone soup %s" % i) loop_epochs.set_description("\nEvolving clone soup %s" % i)
@@ -195,7 +195,7 @@ class SoupSpawnExperiment:
number_clones = number_clones or self.nr_clones number_clones = number_clones or self.nr_clones
df = pd.DataFrame( df = pd.DataFrame(
columns=['parent', 'MAE_pre', 'MAE_post', 'MSE_pre', 'MSE_post', 'MIM_pre', 'MIM_post', 'noise', columns=['name', 'parent', 'MAE_pre', 'MAE_post', 'MSE_pre', 'MSE_post', 'MIM_pre', 'MIM_post', 'noise',
'status_post']) 'status_post'])
# MAE_pre, MSE_pre, MIM_pre = 0, 0, 0 # MAE_pre, MSE_pre, MIM_pre = 0, 0, 0
@@ -232,6 +232,7 @@ class SoupSpawnExperiment:
MIM_pre = mean_invariate_manhattan_distance(net_target_data, clone_pre_weights) MIM_pre = mean_invariate_manhattan_distance(net_target_data, clone_pre_weights)
df.loc[len(df)] = [clone.name, net.name, MAE_pre, 0, MSE_pre, 0, MIM_pre, 0, self.noise, ""] df.loc[len(df)] = [clone.name, net.name, MAE_pre, 0, MSE_pre, 0, MIM_pre, 0, self.noise, ""]
# df.loc[len(df)] = [clone.name, net.name, MAE_pre, 0, 0, 0, 0, 0, self.noise, ""]
net.children.append(clone) net.children.append(clone)
self.clones.append(clone) self.clones.append(clone)
@@ -262,9 +263,14 @@ class SoupSpawnExperiment:
f"\nMSE({i},{j}): {MSE_post}" f"\nMSE({i},{j}): {MSE_post}"
f"\nMAE({i},{j}): {MAE_post}" f"\nMAE({i},{j}): {MAE_post}"
f"\nMIM({i},{j}): {MIM_post}\n") f"\nMIM({i},{j}): {MIM_post}\n")
self.parents_clones_id_functions.append(clone): self.parents_clones_id_functions.append(clone)
df.loc[df.name==clone.name, ["MAE_post", "MSE_post", "MIM_post", "status_post"]] = [MAE_post, MSE_post, MIM_post, clone.is_fixpoint] # df.loc[df.name == clone.name, ["MAE_post", "MSE_post", "MIM_post"]] = [MAE_pre, MSE_pre, MIM_pre]
df.loc[df.name == clone.name, ["MAE_post", "MSE_post", "MIM_post", "status_post"]] = [MAE_post,
MSE_post,
MIM_post,
clone.is_fixpoint]
# Finally take parent net {i} and finish it's training for comparison to clone development. # Finally take parent net {i} and finish it's training for comparison to clone development.
for _ in range(self.epochs - 1): for _ in range(self.epochs - 1):
@@ -289,22 +295,21 @@ class SoupSpawnExperiment:
plot_loss(self.loss_history, self.directory) plot_loss(self.loss_history, self.directory)
if __name__ == "__main__": if __name__ == "__main__":
NET_INPUT_SIZE = 4 NET_INPUT_SIZE = 4
NET_OUT_SIZE = 1 NET_OUT_SIZE = 1
# Define number of runs & name: # Define number of runs & name:
ST_runs = 1 ST_runs = 3
ST_runs_name = "test-27" ST_runs_name = "test-27"
soup_ST_steps = 2500 soup_ST_steps = 1500
soup_epochs = 2 soup_epochs = 2
soup_log_step_size = 10 soup_log_step_size = 10
# Define number of networks & their architecture # Define number of networks & their architecture
nr_clones = 3 nr_clones = 5
soup_population_size = 2 soup_population_size = 3
soup_net_hidden_size = 2 soup_net_hidden_size = 2
soup_net_learning_rate = 0.04 soup_net_learning_rate = 0.04
soup_attack_chance = 10 soup_attack_chance = 10
@@ -312,7 +317,7 @@ if __name__ == "__main__":
print(f"Running the Soup-Spawn experiment:") print(f"Running the Soup-Spawn experiment:")
exp_list = [] exp_list = []
for noise_factor in range(2, 3): for noise_factor in range(2, 5):
exp = SoupSpawnExperiment( exp = SoupSpawnExperiment(
population_size=soup_population_size, population_size=soup_population_size,
log_step_size=soup_log_step_size, log_step_size=soup_log_step_size,
@@ -333,15 +338,28 @@ if __name__ == "__main__":
pickle.dump(exp_list, open(f"{directory}/experiment_pickle_{soup_name_hash}.p", "wb")) pickle.dump(exp_list, open(f"{directory}/experiment_pickle_{soup_name_hash}.p", "wb"))
print(f"\nSaved experiment to {directory}.") print(f"\nSaved experiment to {directory}.")
# Boxplot with counts of nr_fixpoints, nr_other, nr_etc. on y-axis # Concat all dataframes, and add columns depending on where clone weights end up after training (rel. to parent)
df = pd.concat([exp.df for exp in exp_list]) df = pd.concat([exp.df for exp in exp_list])
sns.countplot(data=df, x="noise", hue="status_post") df = df.dropna().reset_index()
plt.savefig(f"output/soup_spawn_basin/{soup_name_hash}/fixpoint_status_countplot.png") df["relative_distance"] = [ (df.loc[i]["MAE_pre"] - df.loc[i]["MAE_post"]) for i in range(len(df))]
df["class"] = ["approaching" if df.loc[i]["relative_distance"] > 0 else "distancing" if df.loc[i]["relative_distance"] < 0 else "stationary" for i in range(len(df))]
# Countplot of all fixpoint clone after training per class. Uncomment and manually adjust xticklabels if x-ax size gets too small.
ax = sns.catplot(kind="count", data=df, x="noise", hue="class", height=5.27, aspect=12.7 / 5.27)
ax.set_axis_labels("Noise Levels", "Clone Fixpoints After Training Count ", fontsize=15)
# ax.set_xticklabels(labels=('10e-10', '10e-9', '10e-8', '10e-7', '10e-6', '10e-5', '10e-4', '10e-3', '10e-2', '10e-1'), fontsize=15)
plt.savefig(f"{directory}/clone_status_after_countplot_{soup_name_hash}.png")
plt.clf()
# Catplot (either kind="point" or "box") that shows before-after training distances to parent # Catplot (either kind="point" or "box") that shows before-after training distances to parent
mlt = df.melt(id_vars=["name", "noise"], value_vars=["MAE_pre", "MAE_post"], var_name="State", value_name="Distance") mlt = df.melt(id_vars=["name", "noise", "class"], value_vars=["MAE_pre", "MAE_post"], var_name="State",
ax = sns.catplot(data=mlt, x="State", y="Distance", col="noise", hue="name", kind="point", col_wrap=min(5, len(exp_list)), sharey=False, legend=False) value_name="Distance")
P = ["blue" if mlt.loc[i]["class"] == "approaching" else "orange" if mlt.loc[i]["class"] == "distancing" else "green" for i in range(len(mlt))]
# P = sns.color_palette(P, as_cmap=False)
ax = sns.catplot(data=mlt, x="State", y="Distance", col="noise", hue="name", kind="point", palette=P,
col_wrap=min(5, len(exp_list)), sharey=False, legend=False)
ax.map(sns.boxplot, "State", "Distance", "noise", linewidth=0.8, order=["MAE_pre", "MAE_post"], whis=[0, 100]) ax.map(sns.boxplot, "State", "Distance", "noise", linewidth=0.8, order=["MAE_pre", "MAE_post"], whis=[0, 100])
ax.set_axis_labels("", "Manhattan Distance To Parent Weights", fontsize=15) ax.set_axis_labels("", "Manhattan Distance To Parent Weights", fontsize=15)
ax.set_xticklabels(labels=('after noise application', 'after training'), fontsize=15) ax.set_xticklabels(labels=('after noise application', 'after training'), fontsize=15)
plt.savefig(f"output/soup_spawn_basin/{soup_name_hash}/clone_distance_catplot.png") plt.savefig(f"{directory}/before_after_distance_catplot_{soup_name_hash}.png")
plt.clf()