mgaughan/mw-lifecycle-analysis
1
0
Fork 0
Code

looking for new phase pca

Browse Source
This commit is contained in:
mgaughan 2025-09-04 10:25:00 -05:00
parent 809e858bbf
commit ccf434db38
4 changed files with 23 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
p2/quest/neurobiber-pca.log
View File

@ -1,9 +1,8 @@
starting the job at: Thu Sep 4 10:09:58 CDT 2025
starting the job at: Thu Sep 4 10:23:23 CDT 2025
setting up the environment
running the neurobiber labeling script
Number of PCs explaining 90% variance: 18
Variance of each PCA component: [88.92832185 39.46471687 32.34601523 20.19544345 14.0083261 11.5837521
7.82584723 6.89064989 6.07988254 5.80726367 5.49782354 4.50587747
4.31482409 2.81997326 2.62989708 2.27205352 2.09396341 2.00076119]
job finished, cleaning up
job pau at: Thu Sep 4 10:10:21 CDT 2025
job pau at: Thu Sep 4 10:23:47 CDT 2025

BIN
p2/quest/phase_090425_biber_kernelpca_affil.png Normal file
View File

Binary file not shown.
Side by Side

After

Width:  |  Height:  |  Size: 797 KiB

28
p2/quest/python_scripts/neurobiber_PCA.py
View File

@ -32,7 +32,7 @@ if __name__ == "__main__":
'''
pca = PCA(n_components=18)
biber_vecs_pca = pca.fit_transform(biber_vecs)
selected_axis = "source"
selected_axis = "phase"
component_variances = np.var(biber_vecs_pca, axis=0)
print("Variance of each PCA component:", component_variances)
@ -41,14 +41,23 @@ if __name__ == "__main__":
le = LabelEncoder()
colors = le.fit_transform(biber_vec_df[selected_axis])
plt.scatter(biber_vecs_pca[:, 0], biber_vecs_pca[:, 1],
c=colors, edgecolor='none', alpha=0.5, cmap="viridis")
plt.xlabel('component 1')
plt.ylabel('component 2')
plt.colorbar()
plot_df = pd.DataFrame({
"PC1": biber_vecs_pca[:, 0],
"PC2": biber_vecs_pca[:, 1],
selected_axis: biber_vec_df[selected_axis].astype(str),
"source":biber_vec_df['source'].astype(str)
})
g = sns.FacetGrid(plot_df, col="source", col_wrap=4, hue=selected_axis, palette="tab10", height=4, sharex=False, sharey=False)
g.map_dataframe(sns.scatterplot, x="PC1", y="PC2", alpha=0.7, s=40)
g.add_legend(title=selected_axis)
g.set_axis_labels("PC1", "PC2")
g.fig.subplots_adjust(top=0.9)
g.fig.suptitle(f"PCA by {selected_axis}, faceted by source")
#plt.savefig("090225_biber_pca_plot.png", dpi=300)
'''
plot_df = pd.DataFrame({
"PC1": biber_vecs_pca[:, 0],
"PC2": biber_vecs_pca[:, 1],
@ -62,6 +71,7 @@ if __name__ == "__main__":
plt.xlabel('component 1')
plt.ylabel('component 2')
plt.legend(title=selected_axis, bbox_to_anchor=(1.05, 1), loc=2)
plt.tight_layout()
plt.savefig(f"{selected_axis}_090425_biber_kernelpca_affil.png", dpi=300)
'''
g.fig.tight_layout()
g.savefig(f"{selected_axis}_090425_biber_kernelpca_affil.png", dpi=300)
plt.show()

2
p2/quest/python_scripts/olmo_parallel_cat.py
View File

@ -14,7 +14,9 @@ from transformers import AutoModelForCausalLM, AutoTokenizer, OlmoForCausalLM
import csv
import pandas as pd
import re
import nltk
nltk.download('punkt')
# ----------------- prompts for LLM
priming = "For the **GIVEN SENTENCE**, please categorize it into one of the defined [[CATEGORIES]]. Each [[CATEGORY]] is described in the TYPOLOGY for reference. Your task is to match the**GIVEN SENTENCE** to the **[[CATEGORY]]** that most accurately describes the content of the comment. Only provide the category as your output. Do not provide any text beyond the category name."