From b54c200dde4ee3bc24a67952cebcf87a6ca31809 Mon Sep 17 00:00:00 2001
From: schne <leo.schneider@ecl19.ec-lyon.fr>
Date: Thu, 4 Apr 2024 18:02:28 +0200
Subject: [PATCH] distance Levenshtein
---
requirements.txt | 3 +++
res_viz.py | 38 +++++++++++++++++++++++++++++++++++++-
2 files changed, 40 insertions(+), 1 deletion(-)
diff --git a/requirements.txt b/requirements.txt
index 46644aa..ad56964 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -10,3 +10,6 @@ seaborn~=0.13.0
pyopenms~=3.1.0
dlomix~=0.0.6
scikit-learn~=1.4.1.post1
+
+Levenshtein~=0.25.0
+keras~=2.15.0
\ No newline at end of file
diff --git a/res_viz.py b/res_viz.py
index 4117e65..9793ecd 100644
--- a/res_viz.py
+++ b/res_viz.py
@@ -3,6 +3,8 @@ import matplotlib
import numpy as np
import matplotlib.pyplot as plt
from dlomix.data import RetentionTimeDataset
+import pandas as pd
+import Levenshtein as lv
epoch = 100
number = 8
BATCH_SIZE=64
@@ -10,6 +12,14 @@ BATCH_SIZE=64
test_rtdata = RetentionTimeDataset(data_source='database/data_holdout.csv',
seq_length=30, batch_size=BATCH_SIZE, test=True)
+rtdata = RetentionTimeDataset(data_source='database/data_train.csv',
+ seq_length=30, batch_size=BATCH_SIZE, val_ratio=0.2, test=False)
+
+data_train = pd.read_csv('database/data_train.csv')
+data_test = pd.read_csv('database/data_holdout.csv')
+train_seq = data_train['sequence']
+test_seq = data_test['sequence']
+
target = test_rtdata.get_split_targets(split="test")
index = random.sample(range(target.size), number)
target_plot = target[index]
@@ -73,4 +83,30 @@ def compare_error():
for i in range(number):
plt.plot(error_plot_1[:,i], c = colors[i], ls = 'dashed')
plt.plot(error_plot_2[:, i], c = colors[i])
- plt.show()
\ No newline at end of file
+ plt.show()
+
+def get_worse_seq(n,epoch):
+ error = error_1[epoch,:]
+ ind = np.argpartition(error, n)[:n]
+ return test_seq[ind].to_list(),ind
+
+def get_nb_iteration(seq_list):
+ nb_list=[]
+ counts = train_seq.value_counts()
+ for seq in seq_list:
+ nb_list.append(counts[seq])
+
+
+def compute_min_distance(seq_list):
+ res=[]
+ for ref in seq_list :
+ min_dist = 10000
+ for seq in train_seq:
+ d = 1 - lv.ratio(seq,ref)
+ if d < min_dist:
+ min_dist = d
+ res.append(min_dist)
+ return res
+
+radom_1000 = index = random.sample(range(test_seq.size), 1000)
+
--
GitLab