File size: 10,202 Bytes
00114fc |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 |
{
"cells": [
{
"cell_type": "code",
"execution_count": 4,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"<class 'pandas.core.frame.DataFrame'>\n",
"RangeIndex: 476 entries, 0 to 475\n",
"Data columns (total 5 columns):\n",
" # Column Non-Null Count Dtype \n",
"--- ------ -------------- ----- \n",
" 0 issuekey 476 non-null object\n",
" 1 created 476 non-null object\n",
" 2 title 476 non-null object\n",
" 3 description 476 non-null object\n",
" 4 storypoint 476 non-null int64 \n",
"dtypes: int64(1), object(4)\n",
"memory usage: 18.7+ KB\n"
]
}
],
"source": [
"import pandas as pd\n",
"project_name = \"APSTUD\"\n",
"\n",
"df = pd.read_csv(\"database\\\\tawos\\\\deep\\\\{}_deep-se.csv\".format(project_name))\n",
"\n",
"df.info()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Pré-Processamento"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {},
"outputs": [],
"source": [
"import re\n",
"from string import punctuation\n",
"\n",
"def escape_tags_and_content(text):\n",
" \"\"\"Escape tags and their content containing text, which is not written in natural language, such as code snippets\"\"\"\n",
"\n",
" NO_TEXT_TAGS = \"code\", \"noformat\"\n",
" for tag in NO_TEXT_TAGS:\n",
" regex_matching_tag = re.compile(\"\\{%s(.*?)\\}(.*?)\\{%s\\}\" % (tag, tag), re.DOTALL)\n",
" text = re.sub(regex_matching_tag, \"\", text)\n",
"\n",
" return text\n",
"\n",
"def escape_tags(text):\n",
" \"\"\"Escape markup tags, but retain their content\"\"\"\n",
"\n",
" ESCAPE_TAGS = \"color\", \"quote\", \"anchor\", \"panel\"\n",
" for tag in ESCAPE_TAGS:\n",
" text = re.sub(\"\\{%s(.*?)\\}\" % tag, \"\", text)\n",
"\n",
" return text\n",
"\n",
"def escape_strings(text):\n",
" \"\"\"Escape line breaks, tabulators, slashes and JIRA heading markup symbols\"\"\"\n",
"\n",
" ESCAPE_STRINGS = \"\\\\r\", \"\\\\n\", \"\\\\t\", \"\\\\f\", \"\\\\v\", \"\\\"\", \"\\\\\\\\\", \"h1. \", \"h2. \", \"h3. \", \"h4. \", \"h5. \", \"h6. \"\n",
" for escape_string in ESCAPE_STRINGS:\n",
" text = text.replace(escape_string, \" \")\n",
"\n",
" return text\n",
"\n",
"def escape_links(text):\n",
" \"\"\"Escape external and internal links, recognized by JIRA markup or leading 'http://' or 'https://' \"\"\"\n",
"\n",
" LINK_STARTERS = r\"\\#\", r\"\\^\", r\"http\\:\\/\\/\", r\"https\\:\\/\\/\", r\"malto\\:\", r\"file\\:\", r\"\\~\"\n",
" for link_starter in LINK_STARTERS:\n",
" text = re.sub(\"\\[(.*?\\\\|)?%s(.*?)\\]\" % link_starter, \"\", text)\n",
" text = re.sub(r\"\\bhttps?://\\S+\", \"\", text)\n",
"\n",
" return text\n",
"\n",
"def escape_hex_character_codes(text):\n",
" \"\"\"Escape characters outside the latin alphabet which are converted to hex code representation\"\"\"\n",
"\n",
" return re.sub(r\"\\\\x\\w\\w\", \"\", text)\n",
"\n",
"def escape_punctuation_boundaries(text):\n",
" \"\"\"Remove all punctuation marks from the beginning and end of words,\n",
" except for trailing period at the end of words\"\"\"\n",
"\n",
" return \" \".join([word.strip(punctuation.replace(\".\", \"\")).lstrip(\".\") for word in text.split()])\n",
"\n",
"def escape_odd_spaces(text):\n",
" \"\"\"Replace several consequent spaces with one space\n",
" and remove spaces from string start and end\"\"\"\n",
"\n",
" text = re.sub(r\"\\s+\", \" \", text)\n",
" text = text.strip()\n",
"\n",
" return text"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Criação do Modelo"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"['models/tawos/APSTUD/model_tawos_APSTUD_tfidf_linear.joblib']"
]
},
"execution_count": 6,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"from sklearn.dummy import DummyRegressor\n",
"from nltk.corpus import stopwords\n",
"from textblob import TextBlob\n",
"import textstat\n",
"from sklearn.feature_extraction.text import TfidfVectorizer\n",
"from sklearn import svm\n",
"from sklearn.linear_model import LinearRegression\n",
"from sklearn.feature_selection import SelectKBest\n",
"import pandas as pd\n",
"from joblib import dump\n",
"\n",
"# carregando os dados\n",
"df = pd.read_csv(\"database\\\\tawos\\\\deep\\\\{}_deep-se.csv\".format(project_name))\n",
"\n",
"# criação de uma nova coluna\n",
"df[\"context\"] = df[\"title\"] + df[\"description\"]\n",
"\n",
"# pré-processamento\n",
"df[\"context\"] = df[\"context\"].apply(lambda x: escape_tags_and_content(x))\n",
"df[\"context\"] = df[\"context\"].apply(lambda x: escape_tags(x))\n",
"df[\"context\"] = df[\"context\"].apply(lambda x: escape_strings(x))\n",
"df[\"context\"] = df[\"context\"].apply(lambda x: escape_links(x))\n",
"df[\"context\"] = df[\"context\"].apply(lambda x: escape_hex_character_codes(x))\n",
"df[\"context\"] = df[\"context\"].apply(lambda x: escape_punctuation_boundaries(x))\n",
"df[\"context\"] = df[\"context\"].apply(lambda x: escape_odd_spaces(x))\n",
"\n",
"# removendo stop-words\n",
"stop = stopwords.words('english')\n",
"df['context'] = df['context'].apply(lambda x: ' '.join([word for word in x.split() if word not in (stop)]))\n",
"\n",
"# renomeando as colunas porque senão dá um problema com a extração de features do NEOSP\n",
"df = df.rename(columns={ \"issuekey\": \"issuekey_\", \"created\": \"created_\", \"description\": \"description_\", \"title\": \"title_\", \"context\": \"context_\", \"storypoint\": \"storypoint_\"})\n",
"y = df[\"storypoint_\"]\n",
"df = df.drop(columns=['storypoint_'])\n",
"\n",
"# 5º coluna -> extração das features para o neosp\n",
"df[\"gunning_fog_\"] = df['context_'].apply(textstat.gunning_fog)\n",
"df[\"flesch_reading_ease_\"] = df['context_'].apply(textstat.flesch_reading_ease)\n",
"df[\"flesch_kincaid_grade_\"] = df['context_'].apply(textstat.flesch_kincaid_grade)\n",
"df[\"smog_index_\"] = df['context_'].apply(textstat.smog_index)\n",
"df[\"coleman_liau_index_\"] = df['context_'].apply(textstat.coleman_liau_index)\n",
"df[\"automated_readability_index_\"] = df['context_'].apply(textstat.automated_readability_index)\n",
"df[\"dale_chall_readability_score_\"] = df['context_'].apply(textstat.dale_chall_readability_score)\n",
"df[\"difficult_words_\"] = df['context_'].apply(textstat.difficult_words)\n",
"df[\"linsear_write_formula_\"] = df['context_'].apply(textstat.linsear_write_formula)\n",
"df[\"polarity_\"] = df[\"context_\"].apply(lambda x: TextBlob(x).sentiment.polarity)\n",
"df[\"subjectivity_\"] = df[\"context_\"].apply(lambda x: TextBlob(x).sentiment.subjectivity)\n",
"# 16º colunas\n",
"\n",
"# Extração das features para o TFIDF\n",
"vectorizer = TfidfVectorizer()\n",
"X_vec = vectorizer.fit_transform(df[\"context_\"])\n",
"#dump(vectorizer, \"vectorizer_tfidf.joblib\")\n",
"dump(vectorizer, \"models/tawos/{}/vectorizer_tawos_{}_tfidf.joblib\".format(project_name, project_name))\n",
"\n",
"df_vec = pd.DataFrame(data = X_vec.toarray(), columns = vectorizer.get_feature_names_out())\n",
"\n",
"# Juntando as features do neosp com o tfidf\n",
"df = df.join(df_vec)\n",
"X = df\n",
"\n",
"############ MbR\n",
"\n",
"model = DummyRegressor(strategy=\"mean\")\n",
"model.fit(X, y)\n",
"#dump(model, \"model_tawos_aloy_mbr.joblib\")\n",
"dump(model, \"models/tawos/{}/model_tawos_{}_mbr.joblib\".format(project_name, project_name))\n",
"\n",
"############ Mediana\n",
"\n",
"model = DummyRegressor(strategy=\"median\")\n",
"model.fit(X, y)\n",
"#dump(model, \"model_tawos_aloy_median.joblib\")\n",
"dump(model, \"models/tawos/{}/model_tawos_{}_median.joblib\".format(project_name, project_name))\n",
"\n",
"########### NEOSP-SVR\n",
"\n",
"model = svm.SVR()\n",
"model.fit(X[X.columns[5:16]], y)\n",
"#dump(model, \"model_tawos_aloy_neosp_svr.joblib\")\n",
"dump(model, \"models/tawos/{}/model_tawos_{}_neosp_svr.joblib\".format(project_name, project_name))\n",
"\n",
"########### NEOSP-LR\n",
"\n",
"model = LinearRegression()\n",
"model.fit(X[X.columns[5:16]], y)\n",
"#dump(model, \"model_tawos_aloy_neosp_linear.joblib\")\n",
"dump(model, \"models/tawos/{}/model_tawos_{}_neosp_linear.joblib\".format(project_name, project_name))\n",
"\n",
"############ TFIDF-SVM\n",
"\n",
"model = svm.SVR()\n",
"model.fit(X[X.columns[16:]], y)\n",
"#dump(model, \"model_tawos_aloy_tfidf_svr.joblib\")\n",
"dump(model, \"models/tawos/{}/model_tawos_{}_tfidf_svr.joblib\".format(project_name, project_name))\n",
"\n",
"############ TFIDF-LR\n",
"\n",
"model = LinearRegression()\n",
"model.fit(X[X.columns[16:]], y)\n",
"#dump(model, \"model_tawos_aloy_tfidf_linear.joblib\")\n",
"dump(model, \"models/tawos/{}/model_tawos_{}_tfidf_linear.joblib\".format(project_name, project_name))"
]
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.10.11"
},
"orig_nbformat": 4
},
"nbformat": 4,
"nbformat_minor": 2
}
|