gpt4all/gpt4all-training/eval_figures.py

29 lines
881 B
Python
Raw Normal View History

2023-03-28 18:47:38 +00:00
import glob
import pickle
import numpy as np
from matplotlib import pyplot as plt
plt.figure()
2023-03-28 20:22:48 +00:00
for fpath in glob.glob('./eval_data/*.pkl'):
2023-03-28 18:47:38 +00:00
parts = fpath.split('__')
2023-04-10 02:14:20 +00:00
model_name = "-".join(fpath.replace(".pkl", "").split("_")[2:])
2023-03-28 18:47:38 +00:00
with open(fpath, 'rb') as f:
data = pickle.load(f)
perplexities = data['perplexities']
perplexities = np.nan_to_num(perplexities, 100)
perplexities = np.clip(perplexities, 0, 100)
2023-04-10 02:14:20 +00:00
if 'alpaca' not in fpath:
identifier = model_name = "-".join(fpath.replace(".pkl", "").split("eval__model-")[1:])
label = 'GPT4all-'
label += identifier
2023-03-28 20:22:48 +00:00
else:
label = 'alpaca-lora'
2023-04-10 02:14:20 +00:00
plt.hist(perplexities, label=label, alpha=.5, bins=50)
2023-03-28 18:47:38 +00:00
plt.xlabel('Perplexity')
plt.ylabel('Frequency')
plt.legend()
plt.savefig('figs/perplexity_hist.png')