Spaces:
Runtime error
Runtime error
File size: 4,196 Bytes
9f7f573 2b9d84c 9f7f573 2b9d84c 9f7f573 90966f7 2b9d84c 9f7f573 2b9d84c |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 |
import pandas as pd
from datasets import load_dataset
import streamlit as st
from clarin_datasets.dataset_to_show import DatasetToShow
class NkjpPosDataset(DatasetToShow):
def __init__(self):
DatasetToShow.__init__(self)
self.data_dict_named = None
self.dataset_name = "clarin-pl/nkjp-pos"
self.description = [
"""
NKJP-POS is a part the National Corpus of Polish (Narodowy Korpus Języka Polskiego).
Its objective is part-of-speech tagging, e.g. nouns, verbs, adjectives, adverbs, etc. During the creation of
corpus, texts of were annotated by humans from various sources, covering many domains and genres.
""",
"Tasks (input, output and metrics)",
"""
Part-of-speech tagging (POS tagging) - tagging words in text with their corresponding part of speech.
Input ('tokens' column): sequence of tokens
Output ('pos_tags' column): sequence of predicted tokens’ classes (35 possible classes, described in detail in the annotation guidelines)
Measurements: F1-score (seqeval)
Example:
Input: ['Zarejestruj', 'się', 'jako', 'bezrobotny', '.']
Input (translated by DeepL): Register as unemployed.
Output: ['impt', 'qub', 'conj', 'subst', 'interp']
"""
]
def load_data(self):
raw_dataset = load_dataset(self.dataset_name)
self.data_dict = {
subset: raw_dataset[subset].to_pandas() for subset in self.subsets
}
self.data_dict_named = {}
for subset in self.subsets:
references = raw_dataset[subset]["pos_tags"]
references_named = [
[
raw_dataset[subset].features["pos_tags"].feature.names[label]
for label in labels
]
for labels in references
]
self.data_dict_named[subset] = pd.DataFrame(
{
"tokens": self.data_dict[subset]["tokens"],
"tags": references_named,
}
)
def show_dataset(self):
header = st.container()
description = st.container()
dataframe_head = st.container()
class_distribution = st.container()
with header:
st.title(self.dataset_name)
with description:
st.header("Dataset description")
st.write(self.description[0])
st.subheader(self.description[1])
st.write(self.description[2])
with dataframe_head:
st.header("First 10 observations of the chosen subset")
subset_to_show = st.selectbox(label="Select subset to see", options=self.subsets)
df_to_show = self.data_dict[subset_to_show].head(10).drop("id", axis="columns")
st.dataframe(df_to_show)
st.text_area(label="LaTeX code", value=df_to_show.style.to_latex())
class_distribution_dict = {}
for subset in self.subsets:
all_labels_from_subset = self.data_dict_named[subset]["tags"].tolist()
all_labels_from_subset = [
x
for subarray in all_labels_from_subset
for x in subarray
]
all_labels_from_subset = pd.Series(all_labels_from_subset)
class_distribution_dict[subset] = (
all_labels_from_subset.value_counts(normalize=True)
.sort_index()
.reset_index()
.rename({"index": "class", 0: subset}, axis="columns")
)
class_distribution_df = pd.merge(
class_distribution_dict["train"],
class_distribution_dict["test"],
on="class",
)
with class_distribution:
st.header("Class distribution in each subset")
st.dataframe(class_distribution_df)
st.text_area(
label="LaTeX code", value=class_distribution_df.style.to_latex()
)
|