ium_z487186/lab_02.ipynb at main

import numpy as np
import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import MinMaxScaler
from collections import Counter

from datasets import load_dataset

dataset = load_dataset("mstz/liver")['train']

Found cached dataset liver (/Users/natalia.szymczyk/.cache/huggingface/datasets/mstz___liver/liver/1.0.0/3115a4001e742dc2c89457a3906d35982a649915f71f35fc5e6d025c786eeacf)
100%|██████████| 1/1 [00:00<00:00, 684.45it/s]

dataset

Dataset({
    features: ['age', 'is_male', 'total_bilirubin', 'direct_ribilubin', 'alkaline_phosphotase', 'alamine_aminotransferasi', 'aspartate_aminotransferase', 'total_proteins', 'albumin', 'albumin_to_globulin_ratio', 'class'],
    num_rows: 583
})

dataset = dataset.to_pandas()

train, test = train_test_split(dataset, test_size=0.2, random_state=42)
train, val = train_test_split(train, test_size=0.2, random_state=42)

train

	age	is_male	total_bilirubin	direct_ribilubin	alkaline_phosphotase	alamine_aminotransferasi	aspartate_aminotransferase	total_proteins	albumin	albumin_to_globulin_ratio	class
107	36	True	0.8	0.2	158	29	39	6.0	2.2	0.50	1
33	38	False	2.6	1.2	410	59	57	5.6	3.0	0.80	1
534	39	True	1.6	0.8	230	88	74	8.0	4.0	1.00	1
204	21	True	0.7	0.2	135	27	26	6.4	3.3	1.00	1
48	32	False	0.6	0.1	176	39	28	6.0	3.0	1.00	0
...	...	...	...	...	...	...	...	...	...	...	...
42	42	True	6.8	3.2	630	25	47	6.1	2.3	0.60	1
179	75	True	8.0	4.6	386	30	25	5.5	1.8	0.48	0
430	53	False	0.7	0.1	182	20	33	4.8	1.9	0.60	0
475	38	True	2.2	1.0	310	119	42	7.9	4.1	1.00	1
425	58	True	0.4	0.1	100	59	126	4.3	2.5	1.40	0

372 rows × 11 columns

val

	age	is_male	total_bilirubin	direct_ribilubin	alkaline_phosphotase	alamine_aminotransferasi	aspartate_aminotransferase	total_proteins	albumin	albumin_to_globulin_ratio	class
582	38	True	1.0	0.3	216	21	24	7.3	4.4	1.50	1
453	46	True	0.7	0.2	224	40	23	7.1	3.0	0.70	0
89	60	True	4.0	1.9	238	119	350	7.1	3.3	0.80	0
71	75	False	0.8	0.2	188	20	29	4.4	1.8	0.60	0
124	28	True	0.6	0.1	177	36	29	6.9	4.1	1.40	1
...	...	...	...	...	...	...	...	...	...	...	...
236	22	True	0.8	0.2	300	57	40	7.9	3.8	0.90	1
487	32	True	0.7	0.2	276	102	190	6.0	2.9	0.93	0
27	34	True	6.2	3.0	240	1680	850	7.2	4.0	1.20	0
307	30	True	0.8	0.2	174	21	47	4.6	2.3	1.00	0
512	56	True	1.0	0.3	195	22	28	5.8	2.6	0.80	1

94 rows × 11 columns

test

	age	is_male	total_bilirubin	direct_ribilubin	alkaline_phosphotase	alamine_aminotransferasi	aspartate_aminotransferase	total_proteins	albumin	albumin_to_globulin_ratio	class
355	19	True	1.4	0.8	178	13	26	8.0	4.6	1.30	1
407	12	True	1.0	0.2	719	157	108	7.2	3.7	1.00	0
90	60	True	5.7	2.8	214	412	850	7.3	3.2	0.78	0
402	42	False	0.5	0.1	162	155	108	8.1	4.0	0.90	0
268	40	True	14.5	6.4	358	50	75	5.7	2.1	0.50	0
...	...	...	...	...	...	...	...	...	...	...	...
516	60	True	0.9	0.3	168	16	24	6.7	3.0	0.80	0
305	54	False	1.4	0.7	195	36	16	7.9	3.7	0.90	1
167	47	False	3.0	1.5	292	64	67	5.6	1.8	0.47	0
312	27	True	1.3	0.6	106	25	54	8.5	4.8	NaN	1
329	21	True	0.7	0.2	211	14	23	7.3	4.1	1.20	1

117 rows × 11 columns

numerical_features = ['age', 'total_bilirubin', 'direct_ribilubin', 'alkaline_phosphotase',
                      'alamine_aminotransferasi', 'aspartate_aminotransferase', 'total_proteins', 'albumin',
                      'albumin_to_globulin_ratio']
train[numerical_features].describe()

	age	total_bilirubin	direct_ribilubin	alkaline_phosphotase	alamine_aminotransferasi	aspartate_aminotransferase	total_proteins	albumin	albumin_to_globulin_ratio
count	372.000000	372.000000	372.000000	372.000000	372.000000	372.000000	372.000000	372.000000	371.000000
mean	44.680108	3.415860	1.494355	286.473118	72.986559	110.147849	6.500269	3.150806	0.959515
std	16.054568	6.736683	2.877245	242.459927	147.472734	306.425153	1.100049	0.806994	0.336514
min	4.000000	0.400000	0.100000	63.000000	10.000000	11.000000	2.700000	0.900000	0.300000
25%	32.000000	0.800000	0.200000	170.000000	24.000000	25.000000	5.775000	2.575000	0.700000
50%	45.000000	1.000000	0.300000	205.500000	35.000000	42.000000	6.600000	3.100000	1.000000
75%	58.000000	2.625000	1.300000	298.000000	60.000000	86.250000	7.200000	3.800000	1.100000
max	85.000000	75.000000	19.700000	2110.000000	1350.000000	4929.000000	9.600000	5.500000	2.800000

label_counter = Counter(train['class'])
print("\nRozkład częstości dla klas:")
for label in label_counter.keys():
    print(f"{label}: {label_counter[label] / len(train) * 100:.2f}%")

Rozkład częstości dla klas:
1: 30.38%
0: 69.62%

scaler = MinMaxScaler()
train[numerical_features] = scaler.fit_transform(train[numerical_features])
test[numerical_features] = scaler.fit_transform(test[numerical_features])
val[numerical_features] = scaler.fit_transform(val[numerical_features])

train

	age	is_male	total_bilirubin	direct_ribilubin	alkaline_phosphotase	alamine_aminotransferasi	aspartate_aminotransferase	total_proteins	albumin	albumin_to_globulin_ratio	class
107	0.395062	True	0.005362	0.005102	0.046409	0.014179	0.005693	0.478261	0.282609	0.080	1
33	0.419753	False	0.029491	0.056122	0.169516	0.036567	0.009353	0.420290	0.456522	0.200	1
534	0.432099	True	0.016086	0.035714	0.081583	0.058209	0.012810	0.768116	0.673913	0.280	1
204	0.209877	True	0.004021	0.005102	0.035173	0.012687	0.003050	0.536232	0.521739	0.280	1
48	0.345679	False	0.002681	0.000000	0.055203	0.021642	0.003457	0.478261	0.456522	0.280	0
...	...	...	...	...	...	...	...	...	...	...	...
42	0.469136	True	0.085791	0.158163	0.276991	0.011194	0.007320	0.492754	0.304348	0.120	1
179	0.876543	True	0.101877	0.229592	0.157792	0.014925	0.002847	0.405797	0.195652	0.072	0
430	0.604938	False	0.004021	0.000000	0.058134	0.007463	0.004473	0.304348	0.217391	0.120	0
475	0.419753	True	0.024129	0.045918	0.120664	0.081343	0.006303	0.753623	0.695652	0.280	1
425	0.666667	True	0.000000	0.000000	0.018075	0.036567	0.023383	0.231884	0.347826	0.440	0

371 rows × 11 columns

train.dropna(inplace=True)
test.dropna(inplace=True)
val.dropna(inplace=True)

train

	age	is_male	total_bilirubin	direct_ribilubin	alkaline_phosphotase	alamine_aminotransferasi	aspartate_aminotransferase	total_proteins	albumin	albumin_to_globulin_ratio	class
107	0.395062	True	0.005362	0.005102	0.046409	0.014179	0.005693	0.478261	0.282609	0.080	1
33	0.419753	False	0.029491	0.056122	0.169516	0.036567	0.009353	0.420290	0.456522	0.200	1
534	0.432099	True	0.016086	0.035714	0.081583	0.058209	0.012810	0.768116	0.673913	0.280	1
204	0.209877	True	0.004021	0.005102	0.035173	0.012687	0.003050	0.536232	0.521739	0.280	1
48	0.345679	False	0.002681	0.000000	0.055203	0.021642	0.003457	0.478261	0.456522	0.280	0
...	...	...	...	...	...	...	...	...	...	...	...
42	0.469136	True	0.085791	0.158163	0.276991	0.011194	0.007320	0.492754	0.304348	0.120	1
179	0.876543	True	0.101877	0.229592	0.157792	0.014925	0.002847	0.405797	0.195652	0.072	0
430	0.604938	False	0.004021	0.000000	0.058134	0.007463	0.004473	0.304348	0.217391	0.120	0
475	0.419753	True	0.024129	0.045918	0.120664	0.081343	0.006303	0.753623	0.695652	0.280	1
425	0.666667	True	0.000000	0.000000	0.018075	0.036567	0.023383	0.231884	0.347826	0.440	0

371 rows × 11 columns

test

	age	is_male	total_bilirubin	direct_ribilubin	alkaline_phosphotase	alamine_aminotransferasi	aspartate_aminotransferase	total_proteins	albumin	albumin_to_globulin_ratio	class
355	0.174419	True	0.030000	0.049645	0.069831	0.002567	0.015385	0.909091	0.947368	0.666667	1
407	0.093023	True	0.016667	0.007092	0.436610	0.187420	0.094231	0.763636	0.710526	0.466667	0
90	0.651163	True	0.173333	0.191489	0.094237	0.514763	0.807692	0.781818	0.578947	0.320000	0
402	0.441860	False	0.000000	0.000000	0.058983	0.184852	0.094231	0.927273	0.789474	0.400000	0
268	0.418605	True	0.466667	0.446809	0.191864	0.050064	0.062500	0.490909	0.289474	0.133333	0
...	...	...	...	...	...	...	...	...	...	...	...
109	0.372093	True	0.013333	0.000000	0.278644	0.017972	0.023077	0.527273	0.473684	0.400000	1
516	0.651163	True	0.013333	0.014184	0.063051	0.006418	0.013462	0.672727	0.526316	0.333333	0
305	0.581395	False	0.030000	0.042553	0.081356	0.032092	0.005769	0.890909	0.710526	0.400000	1
167	0.500000	False	0.083333	0.099291	0.147119	0.068036	0.054808	0.472727	0.210526	0.113333	0
329	0.197674	True	0.006667	0.007092	0.092203	0.003851	0.012500	0.781818	0.815789	0.600000	1

115 rows × 11 columns

val

	age	is_male	total_bilirubin	direct_ribilubin	alkaline_phosphotase	alamine_aminotransferasi	aspartate_aminotransferase	total_proteins	albumin	albumin_to_globulin_ratio	class
582	0.403226	True	0.018100	0.017094	0.069175	0.005528	0.004090	0.725490	0.731707	0.846154	1
453	0.532258	True	0.004525	0.008547	0.074029	0.015075	0.003749	0.686275	0.390244	0.230769	0
89	0.758065	True	0.153846	0.153846	0.082524	0.054774	0.115201	0.686275	0.463415	0.307692	0
71	1.000000	False	0.009050	0.008547	0.052184	0.005025	0.005794	0.156863	0.097561	0.153846	0
124	0.241935	True	0.000000	0.000000	0.045510	0.013065	0.005794	0.647059	0.658537	0.769231	1
...	...	...	...	...	...	...	...	...	...	...	...
236	0.145161	True	0.009050	0.008547	0.120146	0.023618	0.009543	0.843137	0.585366	0.384615	1
487	0.306452	True	0.004525	0.008547	0.105583	0.046231	0.060668	0.470588	0.365854	0.407692	0
27	0.338710	True	0.253394	0.247863	0.083738	0.839196	0.285617	0.705882	0.634146	0.615385	0
307	0.274194	True	0.009050	0.008547	0.043689	0.005528	0.011929	0.196078	0.219512	0.461538	0
512	0.693548	True	0.018100	0.017094	0.056432	0.006030	0.005453	0.431373	0.292683	0.307692	1

93 rows × 11 columns

75 KiB Raw Permalink Blame History Unescape Escape

75 KiB

Raw Permalink Blame History