908 lines
212 KiB
Text
908 lines
212 KiB
Text
{
|
|
"cells": [
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {},
|
|
"source": [
|
|
"# Train a field of study (domain) classification of sentences on the [MAG](https://github.com/allenai/scibert/tree/master/data/text_classification/mag) dataset\n",
|
|
"\n",
|
|
"[SciBERT achieves an F1-score of 0.6571 on this dataset.](https://paperswithcode.com/sota/sentence-classification-on-paper-field) \n",
|
|
"This notebook shows that better results can be achieved without even using transformers."
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 1,
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stderr",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"100%|██████████| 84000/84000 [00:16<00:00, 4964.08it/s]\n",
|
|
"100%|██████████| 22399/22399 [00:05<00:00, 3847.04it/s]\n"
|
|
]
|
|
}
|
|
],
|
|
"source": [
|
|
"import json\n",
|
|
"from typing import Tuple\n",
|
|
"from great_ai.utilities import clean, simple_parallel_map\n",
|
|
"from tqdm.cli import tqdm\n",
|
|
"\n",
|
|
"\n",
|
|
"def preprocess(line: str) -> Tuple[str, str]:\n",
|
|
" data_point = json.loads(line)\n",
|
|
"\n",
|
|
" return (clean(data_point[\"text\"], convert_to_ascii=True), data_point[\"label\"])\n",
|
|
"\n",
|
|
"\n",
|
|
"with open(\"../../tutorial/data/train.txt\", encoding=\"utf-8\") as f:\n",
|
|
" training_data = simple_parallel_map(preprocess, f.readlines())\n",
|
|
"\n",
|
|
"X_train = [d[0] for d in training_data]\n",
|
|
"y_train = [d[1] for d in training_data]\n",
|
|
"\n",
|
|
"\n",
|
|
"with open(\"../../tutorial/data/test.txt\", encoding=\"utf-8\") as f:\n",
|
|
" test_data = simple_parallel_map(preprocess, f.readlines())\n",
|
|
"\n",
|
|
"X_test = [d[0] for d in test_data]\n",
|
|
"y_test = [d[1] for d in test_data]"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 2,
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"data": {
|
|
"image/png": "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",
|
|
"text/plain": [
|
|
"<Figure size 576x360 with 1 Axes>"
|
|
]
|
|
},
|
|
"metadata": {},
|
|
"output_type": "display_data"
|
|
}
|
|
],
|
|
"source": [
|
|
"from collections import Counter\n",
|
|
"import matplotlib.pyplot as plt\n",
|
|
"\n",
|
|
"domains, counts = zip(*Counter(y_train).most_common())\n",
|
|
"\n",
|
|
"# Configure matplotlib to have nice, high-resolution charts\n",
|
|
"%matplotlib inline\n",
|
|
"plt.rcParams[\"figure.facecolor\"] = \"white\"\n",
|
|
"plt.rcParams[\"font.size\"] = 18\n",
|
|
"plt.rcParams[\"figure.figsize\"] = (8, 5)\n",
|
|
"\n",
|
|
"fig, ax = plt.subplots()\n",
|
|
"\n",
|
|
"plt.xticks(rotation=90)\n",
|
|
"ax.bar(domains, counts)\n",
|
|
"ax.set_ylabel(\"Count\")\n",
|
|
"fig.tight_layout()\n",
|
|
"fig.savefig(\"mag-distribution.png\", dpi=500)\n",
|
|
"None"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 3,
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"from sklearn.naive_bayes import MultinomialNB\n",
|
|
"from sklearn.pipeline import Pipeline\n",
|
|
"from sklearn.feature_extraction.text import TfidfVectorizer\n",
|
|
"\n",
|
|
"\n",
|
|
"def create_pipeline() -> Pipeline:\n",
|
|
" return Pipeline(\n",
|
|
" steps=[\n",
|
|
" (\"vectorizer\", TfidfVectorizer(sublinear_tf=True)),\n",
|
|
" (\"classifier\", MultinomialNB()),\n",
|
|
" ]\n",
|
|
" )"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 4,
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"Fitting 3 folds for each of 24 candidates, totalling 72 fits\n"
|
|
]
|
|
},
|
|
{
|
|
"data": {
|
|
"text/html": [
|
|
"<div>\n",
|
|
"<style scoped>\n",
|
|
" .dataframe tbody tr th:only-of-type {\n",
|
|
" vertical-align: middle;\n",
|
|
" }\n",
|
|
"\n",
|
|
" .dataframe tbody tr th {\n",
|
|
" vertical-align: top;\n",
|
|
" }\n",
|
|
"\n",
|
|
" .dataframe thead th {\n",
|
|
" text-align: right;\n",
|
|
" }\n",
|
|
"</style>\n",
|
|
"<table border=\"1\" class=\"dataframe\">\n",
|
|
" <thead>\n",
|
|
" <tr style=\"text-align: right;\">\n",
|
|
" <th></th>\n",
|
|
" <th>mean_fit_time</th>\n",
|
|
" <th>std_fit_time</th>\n",
|
|
" <th>mean_score_time</th>\n",
|
|
" <th>std_score_time</th>\n",
|
|
" <th>param_classifier__alpha</th>\n",
|
|
" <th>param_classifier__fit_prior</th>\n",
|
|
" <th>param_vectorizer__max_df</th>\n",
|
|
" <th>param_vectorizer__min_df</th>\n",
|
|
" <th>params</th>\n",
|
|
" <th>split0_test_score</th>\n",
|
|
" <th>split1_test_score</th>\n",
|
|
" <th>split2_test_score</th>\n",
|
|
" <th>mean_test_score</th>\n",
|
|
" <th>std_test_score</th>\n",
|
|
" <th>rank_test_score</th>\n",
|
|
" </tr>\n",
|
|
" </thead>\n",
|
|
" <tbody>\n",
|
|
" <tr>\n",
|
|
" <th>12</th>\n",
|
|
" <td>2.352721</td>\n",
|
|
" <td>0.257936</td>\n",
|
|
" <td>1.357399</td>\n",
|
|
" <td>0.094580</td>\n",
|
|
" <td>1</td>\n",
|
|
" <td>True</td>\n",
|
|
" <td>0.05</td>\n",
|
|
" <td>5</td>\n",
|
|
" <td>{'classifier__alpha': 1, 'classifier__fit_prio...</td>\n",
|
|
" <td>0.677283</td>\n",
|
|
" <td>0.670398</td>\n",
|
|
" <td>0.665671</td>\n",
|
|
" <td>0.671118</td>\n",
|
|
" <td>0.004768</td>\n",
|
|
" <td>1</td>\n",
|
|
" </tr>\n",
|
|
" <tr>\n",
|
|
" <th>15</th>\n",
|
|
" <td>2.544958</td>\n",
|
|
" <td>0.051976</td>\n",
|
|
" <td>1.245658</td>\n",
|
|
" <td>0.137108</td>\n",
|
|
" <td>1</td>\n",
|
|
" <td>True</td>\n",
|
|
" <td>0.1</td>\n",
|
|
" <td>5</td>\n",
|
|
" <td>{'classifier__alpha': 1, 'classifier__fit_prio...</td>\n",
|
|
" <td>0.676970</td>\n",
|
|
" <td>0.670244</td>\n",
|
|
" <td>0.666123</td>\n",
|
|
" <td>0.671112</td>\n",
|
|
" <td>0.004470</td>\n",
|
|
" <td>2</td>\n",
|
|
" </tr>\n",
|
|
" <tr>\n",
|
|
" <th>18</th>\n",
|
|
" <td>2.472888</td>\n",
|
|
" <td>0.053978</td>\n",
|
|
" <td>1.447181</td>\n",
|
|
" <td>0.051170</td>\n",
|
|
" <td>1</td>\n",
|
|
" <td>False</td>\n",
|
|
" <td>0.05</td>\n",
|
|
" <td>5</td>\n",
|
|
" <td>{'classifier__alpha': 1, 'classifier__fit_prio...</td>\n",
|
|
" <td>0.675678</td>\n",
|
|
" <td>0.669321</td>\n",
|
|
" <td>0.665305</td>\n",
|
|
" <td>0.670101</td>\n",
|
|
" <td>0.004271</td>\n",
|
|
" <td>3</td>\n",
|
|
" </tr>\n",
|
|
" <tr>\n",
|
|
" <th>21</th>\n",
|
|
" <td>2.467637</td>\n",
|
|
" <td>0.005358</td>\n",
|
|
" <td>1.216254</td>\n",
|
|
" <td>0.048319</td>\n",
|
|
" <td>1</td>\n",
|
|
" <td>False</td>\n",
|
|
" <td>0.1</td>\n",
|
|
" <td>5</td>\n",
|
|
" <td>{'classifier__alpha': 1, 'classifier__fit_prio...</td>\n",
|
|
" <td>0.675569</td>\n",
|
|
" <td>0.669482</td>\n",
|
|
" <td>0.665225</td>\n",
|
|
" <td>0.670092</td>\n",
|
|
" <td>0.004245</td>\n",
|
|
" <td>4</td>\n",
|
|
" </tr>\n",
|
|
" <tr>\n",
|
|
" <th>3</th>\n",
|
|
" <td>2.477612</td>\n",
|
|
" <td>0.134880</td>\n",
|
|
" <td>1.428521</td>\n",
|
|
" <td>0.017417</td>\n",
|
|
" <td>0.5</td>\n",
|
|
" <td>True</td>\n",
|
|
" <td>0.1</td>\n",
|
|
" <td>5</td>\n",
|
|
" <td>{'classifier__alpha': 0.5, 'classifier__fit_pr...</td>\n",
|
|
" <td>0.675225</td>\n",
|
|
" <td>0.668912</td>\n",
|
|
" <td>0.665652</td>\n",
|
|
" <td>0.669930</td>\n",
|
|
" <td>0.003974</td>\n",
|
|
" <td>5</td>\n",
|
|
" </tr>\n",
|
|
" <tr>\n",
|
|
" <th>0</th>\n",
|
|
" <td>2.484810</td>\n",
|
|
" <td>0.067836</td>\n",
|
|
" <td>1.249979</td>\n",
|
|
" <td>0.209959</td>\n",
|
|
" <td>0.5</td>\n",
|
|
" <td>True</td>\n",
|
|
" <td>0.05</td>\n",
|
|
" <td>5</td>\n",
|
|
" <td>{'classifier__alpha': 0.5, 'classifier__fit_pr...</td>\n",
|
|
" <td>0.675264</td>\n",
|
|
" <td>0.668781</td>\n",
|
|
" <td>0.665394</td>\n",
|
|
" <td>0.669813</td>\n",
|
|
" <td>0.004095</td>\n",
|
|
" <td>6</td>\n",
|
|
" </tr>\n",
|
|
" <tr>\n",
|
|
" <th>9</th>\n",
|
|
" <td>2.452453</td>\n",
|
|
" <td>0.049376</td>\n",
|
|
" <td>1.279694</td>\n",
|
|
" <td>0.089200</td>\n",
|
|
" <td>0.5</td>\n",
|
|
" <td>False</td>\n",
|
|
" <td>0.1</td>\n",
|
|
" <td>5</td>\n",
|
|
" <td>{'classifier__alpha': 0.5, 'classifier__fit_pr...</td>\n",
|
|
" <td>0.674199</td>\n",
|
|
" <td>0.668005</td>\n",
|
|
" <td>0.664456</td>\n",
|
|
" <td>0.668887</td>\n",
|
|
" <td>0.004026</td>\n",
|
|
" <td>7</td>\n",
|
|
" </tr>\n",
|
|
" <tr>\n",
|
|
" <th>6</th>\n",
|
|
" <td>2.748628</td>\n",
|
|
" <td>0.098077</td>\n",
|
|
" <td>1.255213</td>\n",
|
|
" <td>0.101334</td>\n",
|
|
" <td>0.5</td>\n",
|
|
" <td>False</td>\n",
|
|
" <td>0.05</td>\n",
|
|
" <td>5</td>\n",
|
|
" <td>{'classifier__alpha': 0.5, 'classifier__fit_pr...</td>\n",
|
|
" <td>0.673996</td>\n",
|
|
" <td>0.668048</td>\n",
|
|
" <td>0.664572</td>\n",
|
|
" <td>0.668872</td>\n",
|
|
" <td>0.003891</td>\n",
|
|
" <td>8</td>\n",
|
|
" </tr>\n",
|
|
" <tr>\n",
|
|
" <th>13</th>\n",
|
|
" <td>2.371907</td>\n",
|
|
" <td>0.204820</td>\n",
|
|
" <td>1.375143</td>\n",
|
|
" <td>0.193984</td>\n",
|
|
" <td>1</td>\n",
|
|
" <td>True</td>\n",
|
|
" <td>0.05</td>\n",
|
|
" <td>20</td>\n",
|
|
" <td>{'classifier__alpha': 1, 'classifier__fit_prio...</td>\n",
|
|
" <td>0.656459</td>\n",
|
|
" <td>0.647433</td>\n",
|
|
" <td>0.644864</td>\n",
|
|
" <td>0.649585</td>\n",
|
|
" <td>0.004972</td>\n",
|
|
" <td>9</td>\n",
|
|
" </tr>\n",
|
|
" <tr>\n",
|
|
" <th>16</th>\n",
|
|
" <td>2.557486</td>\n",
|
|
" <td>0.184332</td>\n",
|
|
" <td>1.272137</td>\n",
|
|
" <td>0.133980</td>\n",
|
|
" <td>1</td>\n",
|
|
" <td>True</td>\n",
|
|
" <td>0.1</td>\n",
|
|
" <td>20</td>\n",
|
|
" <td>{'classifier__alpha': 1, 'classifier__fit_prio...</td>\n",
|
|
" <td>0.656196</td>\n",
|
|
" <td>0.646960</td>\n",
|
|
" <td>0.645436</td>\n",
|
|
" <td>0.649530</td>\n",
|
|
" <td>0.004754</td>\n",
|
|
" <td>10</td>\n",
|
|
" </tr>\n",
|
|
" <tr>\n",
|
|
" <th>4</th>\n",
|
|
" <td>2.362093</td>\n",
|
|
" <td>0.109796</td>\n",
|
|
" <td>1.344960</td>\n",
|
|
" <td>0.083562</td>\n",
|
|
" <td>0.5</td>\n",
|
|
" <td>True</td>\n",
|
|
" <td>0.1</td>\n",
|
|
" <td>20</td>\n",
|
|
" <td>{'classifier__alpha': 0.5, 'classifier__fit_pr...</td>\n",
|
|
" <td>0.654424</td>\n",
|
|
" <td>0.644690</td>\n",
|
|
" <td>0.644991</td>\n",
|
|
" <td>0.648035</td>\n",
|
|
" <td>0.004519</td>\n",
|
|
" <td>11</td>\n",
|
|
" </tr>\n",
|
|
" <tr>\n",
|
|
" <th>1</th>\n",
|
|
" <td>2.668012</td>\n",
|
|
" <td>0.131751</td>\n",
|
|
" <td>1.383441</td>\n",
|
|
" <td>0.049157</td>\n",
|
|
" <td>0.5</td>\n",
|
|
" <td>True</td>\n",
|
|
" <td>0.05</td>\n",
|
|
" <td>20</td>\n",
|
|
" <td>{'classifier__alpha': 0.5, 'classifier__fit_pr...</td>\n",
|
|
" <td>0.654622</td>\n",
|
|
" <td>0.644901</td>\n",
|
|
" <td>0.644425</td>\n",
|
|
" <td>0.647983</td>\n",
|
|
" <td>0.004699</td>\n",
|
|
" <td>12</td>\n",
|
|
" </tr>\n",
|
|
" <tr>\n",
|
|
" <th>19</th>\n",
|
|
" <td>2.517576</td>\n",
|
|
" <td>0.154192</td>\n",
|
|
" <td>1.374269</td>\n",
|
|
" <td>0.026773</td>\n",
|
|
" <td>1</td>\n",
|
|
" <td>False</td>\n",
|
|
" <td>0.05</td>\n",
|
|
" <td>20</td>\n",
|
|
" <td>{'classifier__alpha': 1, 'classifier__fit_prio...</td>\n",
|
|
" <td>0.655125</td>\n",
|
|
" <td>0.645586</td>\n",
|
|
" <td>0.643166</td>\n",
|
|
" <td>0.647959</td>\n",
|
|
" <td>0.005163</td>\n",
|
|
" <td>13</td>\n",
|
|
" </tr>\n",
|
|
" <tr>\n",
|
|
" <th>22</th>\n",
|
|
" <td>2.386240</td>\n",
|
|
" <td>0.069718</td>\n",
|
|
" <td>0.967638</td>\n",
|
|
" <td>0.076766</td>\n",
|
|
" <td>1</td>\n",
|
|
" <td>False</td>\n",
|
|
" <td>0.1</td>\n",
|
|
" <td>20</td>\n",
|
|
" <td>{'classifier__alpha': 1, 'classifier__fit_prio...</td>\n",
|
|
" <td>0.654254</td>\n",
|
|
" <td>0.645409</td>\n",
|
|
" <td>0.643514</td>\n",
|
|
" <td>0.647726</td>\n",
|
|
" <td>0.004680</td>\n",
|
|
" <td>14</td>\n",
|
|
" </tr>\n",
|
|
" <tr>\n",
|
|
" <th>7</th>\n",
|
|
" <td>2.416211</td>\n",
|
|
" <td>0.138001</td>\n",
|
|
" <td>1.316785</td>\n",
|
|
" <td>0.036402</td>\n",
|
|
" <td>0.5</td>\n",
|
|
" <td>False</td>\n",
|
|
" <td>0.05</td>\n",
|
|
" <td>20</td>\n",
|
|
" <td>{'classifier__alpha': 0.5, 'classifier__fit_pr...</td>\n",
|
|
" <td>0.652727</td>\n",
|
|
" <td>0.643691</td>\n",
|
|
" <td>0.642749</td>\n",
|
|
" <td>0.646389</td>\n",
|
|
" <td>0.004498</td>\n",
|
|
" <td>15</td>\n",
|
|
" </tr>\n",
|
|
" <tr>\n",
|
|
" <th>10</th>\n",
|
|
" <td>2.397768</td>\n",
|
|
" <td>0.244951</td>\n",
|
|
" <td>1.330007</td>\n",
|
|
" <td>0.008983</td>\n",
|
|
" <td>0.5</td>\n",
|
|
" <td>False</td>\n",
|
|
" <td>0.1</td>\n",
|
|
" <td>20</td>\n",
|
|
" <td>{'classifier__alpha': 0.5, 'classifier__fit_pr...</td>\n",
|
|
" <td>0.652290</td>\n",
|
|
" <td>0.642975</td>\n",
|
|
" <td>0.643421</td>\n",
|
|
" <td>0.646229</td>\n",
|
|
" <td>0.004290</td>\n",
|
|
" <td>16</td>\n",
|
|
" </tr>\n",
|
|
" <tr>\n",
|
|
" <th>14</th>\n",
|
|
" <td>2.590127</td>\n",
|
|
" <td>0.054396</td>\n",
|
|
" <td>1.155976</td>\n",
|
|
" <td>0.187602</td>\n",
|
|
" <td>1</td>\n",
|
|
" <td>True</td>\n",
|
|
" <td>0.05</td>\n",
|
|
" <td>100</td>\n",
|
|
" <td>{'classifier__alpha': 1, 'classifier__fit_prio...</td>\n",
|
|
" <td>0.587097</td>\n",
|
|
" <td>0.577411</td>\n",
|
|
" <td>0.580170</td>\n",
|
|
" <td>0.581560</td>\n",
|
|
" <td>0.004075</td>\n",
|
|
" <td>17</td>\n",
|
|
" </tr>\n",
|
|
" <tr>\n",
|
|
" <th>17</th>\n",
|
|
" <td>2.497556</td>\n",
|
|
" <td>0.085952</td>\n",
|
|
" <td>1.169808</td>\n",
|
|
" <td>0.061336</td>\n",
|
|
" <td>1</td>\n",
|
|
" <td>True</td>\n",
|
|
" <td>0.1</td>\n",
|
|
" <td>100</td>\n",
|
|
" <td>{'classifier__alpha': 1, 'classifier__fit_prio...</td>\n",
|
|
" <td>0.585713</td>\n",
|
|
" <td>0.577197</td>\n",
|
|
" <td>0.580763</td>\n",
|
|
" <td>0.581224</td>\n",
|
|
" <td>0.003492</td>\n",
|
|
" <td>18</td>\n",
|
|
" </tr>\n",
|
|
" <tr>\n",
|
|
" <th>2</th>\n",
|
|
" <td>2.467611</td>\n",
|
|
" <td>0.168399</td>\n",
|
|
" <td>1.313936</td>\n",
|
|
" <td>0.109695</td>\n",
|
|
" <td>0.5</td>\n",
|
|
" <td>True</td>\n",
|
|
" <td>0.05</td>\n",
|
|
" <td>100</td>\n",
|
|
" <td>{'classifier__alpha': 0.5, 'classifier__fit_pr...</td>\n",
|
|
" <td>0.586346</td>\n",
|
|
" <td>0.576987</td>\n",
|
|
" <td>0.579852</td>\n",
|
|
" <td>0.581062</td>\n",
|
|
" <td>0.003915</td>\n",
|
|
" <td>19</td>\n",
|
|
" </tr>\n",
|
|
" <tr>\n",
|
|
" <th>5</th>\n",
|
|
" <td>2.339322</td>\n",
|
|
" <td>0.092595</td>\n",
|
|
" <td>1.209515</td>\n",
|
|
" <td>0.132825</td>\n",
|
|
" <td>0.5</td>\n",
|
|
" <td>True</td>\n",
|
|
" <td>0.1</td>\n",
|
|
" <td>100</td>\n",
|
|
" <td>{'classifier__alpha': 0.5, 'classifier__fit_pr...</td>\n",
|
|
" <td>0.585634</td>\n",
|
|
" <td>0.576217</td>\n",
|
|
" <td>0.580320</td>\n",
|
|
" <td>0.580723</td>\n",
|
|
" <td>0.003855</td>\n",
|
|
" <td>20</td>\n",
|
|
" </tr>\n",
|
|
" <tr>\n",
|
|
" <th>20</th>\n",
|
|
" <td>2.644488</td>\n",
|
|
" <td>0.119967</td>\n",
|
|
" <td>1.238554</td>\n",
|
|
" <td>0.048871</td>\n",
|
|
" <td>1</td>\n",
|
|
" <td>False</td>\n",
|
|
" <td>0.05</td>\n",
|
|
" <td>100</td>\n",
|
|
" <td>{'classifier__alpha': 1, 'classifier__fit_prio...</td>\n",
|
|
" <td>0.581096</td>\n",
|
|
" <td>0.572833</td>\n",
|
|
" <td>0.576280</td>\n",
|
|
" <td>0.576736</td>\n",
|
|
" <td>0.003389</td>\n",
|
|
" <td>21</td>\n",
|
|
" </tr>\n",
|
|
" <tr>\n",
|
|
" <th>8</th>\n",
|
|
" <td>2.456030</td>\n",
|
|
" <td>0.072136</td>\n",
|
|
" <td>1.240463</td>\n",
|
|
" <td>0.092383</td>\n",
|
|
" <td>0.5</td>\n",
|
|
" <td>False</td>\n",
|
|
" <td>0.05</td>\n",
|
|
" <td>100</td>\n",
|
|
" <td>{'classifier__alpha': 0.5, 'classifier__fit_pr...</td>\n",
|
|
" <td>0.580841</td>\n",
|
|
" <td>0.572391</td>\n",
|
|
" <td>0.575731</td>\n",
|
|
" <td>0.576321</td>\n",
|
|
" <td>0.003475</td>\n",
|
|
" <td>22</td>\n",
|
|
" </tr>\n",
|
|
" <tr>\n",
|
|
" <th>23</th>\n",
|
|
" <td>2.335796</td>\n",
|
|
" <td>0.082338</td>\n",
|
|
" <td>0.786187</td>\n",
|
|
" <td>0.073499</td>\n",
|
|
" <td>1</td>\n",
|
|
" <td>False</td>\n",
|
|
" <td>0.1</td>\n",
|
|
" <td>100</td>\n",
|
|
" <td>{'classifier__alpha': 1, 'classifier__fit_prio...</td>\n",
|
|
" <td>0.579769</td>\n",
|
|
" <td>0.572373</td>\n",
|
|
" <td>0.576699</td>\n",
|
|
" <td>0.576280</td>\n",
|
|
" <td>0.003034</td>\n",
|
|
" <td>23</td>\n",
|
|
" </tr>\n",
|
|
" <tr>\n",
|
|
" <th>11</th>\n",
|
|
" <td>2.530331</td>\n",
|
|
" <td>0.107199</td>\n",
|
|
" <td>1.247842</td>\n",
|
|
" <td>0.078229</td>\n",
|
|
" <td>0.5</td>\n",
|
|
" <td>False</td>\n",
|
|
" <td>0.1</td>\n",
|
|
" <td>100</td>\n",
|
|
" <td>{'classifier__alpha': 0.5, 'classifier__fit_pr...</td>\n",
|
|
" <td>0.579743</td>\n",
|
|
" <td>0.571607</td>\n",
|
|
" <td>0.576352</td>\n",
|
|
" <td>0.575901</td>\n",
|
|
" <td>0.003337</td>\n",
|
|
" <td>24</td>\n",
|
|
" </tr>\n",
|
|
" </tbody>\n",
|
|
"</table>\n",
|
|
"</div>"
|
|
],
|
|
"text/plain": [
|
|
" mean_fit_time std_fit_time mean_score_time std_score_time \\\n",
|
|
"12 2.352721 0.257936 1.357399 0.094580 \n",
|
|
"15 2.544958 0.051976 1.245658 0.137108 \n",
|
|
"18 2.472888 0.053978 1.447181 0.051170 \n",
|
|
"21 2.467637 0.005358 1.216254 0.048319 \n",
|
|
"3 2.477612 0.134880 1.428521 0.017417 \n",
|
|
"0 2.484810 0.067836 1.249979 0.209959 \n",
|
|
"9 2.452453 0.049376 1.279694 0.089200 \n",
|
|
"6 2.748628 0.098077 1.255213 0.101334 \n",
|
|
"13 2.371907 0.204820 1.375143 0.193984 \n",
|
|
"16 2.557486 0.184332 1.272137 0.133980 \n",
|
|
"4 2.362093 0.109796 1.344960 0.083562 \n",
|
|
"1 2.668012 0.131751 1.383441 0.049157 \n",
|
|
"19 2.517576 0.154192 1.374269 0.026773 \n",
|
|
"22 2.386240 0.069718 0.967638 0.076766 \n",
|
|
"7 2.416211 0.138001 1.316785 0.036402 \n",
|
|
"10 2.397768 0.244951 1.330007 0.008983 \n",
|
|
"14 2.590127 0.054396 1.155976 0.187602 \n",
|
|
"17 2.497556 0.085952 1.169808 0.061336 \n",
|
|
"2 2.467611 0.168399 1.313936 0.109695 \n",
|
|
"5 2.339322 0.092595 1.209515 0.132825 \n",
|
|
"20 2.644488 0.119967 1.238554 0.048871 \n",
|
|
"8 2.456030 0.072136 1.240463 0.092383 \n",
|
|
"23 2.335796 0.082338 0.786187 0.073499 \n",
|
|
"11 2.530331 0.107199 1.247842 0.078229 \n",
|
|
"\n",
|
|
" param_classifier__alpha param_classifier__fit_prior \\\n",
|
|
"12 1 True \n",
|
|
"15 1 True \n",
|
|
"18 1 False \n",
|
|
"21 1 False \n",
|
|
"3 0.5 True \n",
|
|
"0 0.5 True \n",
|
|
"9 0.5 False \n",
|
|
"6 0.5 False \n",
|
|
"13 1 True \n",
|
|
"16 1 True \n",
|
|
"4 0.5 True \n",
|
|
"1 0.5 True \n",
|
|
"19 1 False \n",
|
|
"22 1 False \n",
|
|
"7 0.5 False \n",
|
|
"10 0.5 False \n",
|
|
"14 1 True \n",
|
|
"17 1 True \n",
|
|
"2 0.5 True \n",
|
|
"5 0.5 True \n",
|
|
"20 1 False \n",
|
|
"8 0.5 False \n",
|
|
"23 1 False \n",
|
|
"11 0.5 False \n",
|
|
"\n",
|
|
" param_vectorizer__max_df param_vectorizer__min_df \\\n",
|
|
"12 0.05 5 \n",
|
|
"15 0.1 5 \n",
|
|
"18 0.05 5 \n",
|
|
"21 0.1 5 \n",
|
|
"3 0.1 5 \n",
|
|
"0 0.05 5 \n",
|
|
"9 0.1 5 \n",
|
|
"6 0.05 5 \n",
|
|
"13 0.05 20 \n",
|
|
"16 0.1 20 \n",
|
|
"4 0.1 20 \n",
|
|
"1 0.05 20 \n",
|
|
"19 0.05 20 \n",
|
|
"22 0.1 20 \n",
|
|
"7 0.05 20 \n",
|
|
"10 0.1 20 \n",
|
|
"14 0.05 100 \n",
|
|
"17 0.1 100 \n",
|
|
"2 0.05 100 \n",
|
|
"5 0.1 100 \n",
|
|
"20 0.05 100 \n",
|
|
"8 0.05 100 \n",
|
|
"23 0.1 100 \n",
|
|
"11 0.1 100 \n",
|
|
"\n",
|
|
" params split0_test_score \\\n",
|
|
"12 {'classifier__alpha': 1, 'classifier__fit_prio... 0.677283 \n",
|
|
"15 {'classifier__alpha': 1, 'classifier__fit_prio... 0.676970 \n",
|
|
"18 {'classifier__alpha': 1, 'classifier__fit_prio... 0.675678 \n",
|
|
"21 {'classifier__alpha': 1, 'classifier__fit_prio... 0.675569 \n",
|
|
"3 {'classifier__alpha': 0.5, 'classifier__fit_pr... 0.675225 \n",
|
|
"0 {'classifier__alpha': 0.5, 'classifier__fit_pr... 0.675264 \n",
|
|
"9 {'classifier__alpha': 0.5, 'classifier__fit_pr... 0.674199 \n",
|
|
"6 {'classifier__alpha': 0.5, 'classifier__fit_pr... 0.673996 \n",
|
|
"13 {'classifier__alpha': 1, 'classifier__fit_prio... 0.656459 \n",
|
|
"16 {'classifier__alpha': 1, 'classifier__fit_prio... 0.656196 \n",
|
|
"4 {'classifier__alpha': 0.5, 'classifier__fit_pr... 0.654424 \n",
|
|
"1 {'classifier__alpha': 0.5, 'classifier__fit_pr... 0.654622 \n",
|
|
"19 {'classifier__alpha': 1, 'classifier__fit_prio... 0.655125 \n",
|
|
"22 {'classifier__alpha': 1, 'classifier__fit_prio... 0.654254 \n",
|
|
"7 {'classifier__alpha': 0.5, 'classifier__fit_pr... 0.652727 \n",
|
|
"10 {'classifier__alpha': 0.5, 'classifier__fit_pr... 0.652290 \n",
|
|
"14 {'classifier__alpha': 1, 'classifier__fit_prio... 0.587097 \n",
|
|
"17 {'classifier__alpha': 1, 'classifier__fit_prio... 0.585713 \n",
|
|
"2 {'classifier__alpha': 0.5, 'classifier__fit_pr... 0.586346 \n",
|
|
"5 {'classifier__alpha': 0.5, 'classifier__fit_pr... 0.585634 \n",
|
|
"20 {'classifier__alpha': 1, 'classifier__fit_prio... 0.581096 \n",
|
|
"8 {'classifier__alpha': 0.5, 'classifier__fit_pr... 0.580841 \n",
|
|
"23 {'classifier__alpha': 1, 'classifier__fit_prio... 0.579769 \n",
|
|
"11 {'classifier__alpha': 0.5, 'classifier__fit_pr... 0.579743 \n",
|
|
"\n",
|
|
" split1_test_score split2_test_score mean_test_score std_test_score \\\n",
|
|
"12 0.670398 0.665671 0.671118 0.004768 \n",
|
|
"15 0.670244 0.666123 0.671112 0.004470 \n",
|
|
"18 0.669321 0.665305 0.670101 0.004271 \n",
|
|
"21 0.669482 0.665225 0.670092 0.004245 \n",
|
|
"3 0.668912 0.665652 0.669930 0.003974 \n",
|
|
"0 0.668781 0.665394 0.669813 0.004095 \n",
|
|
"9 0.668005 0.664456 0.668887 0.004026 \n",
|
|
"6 0.668048 0.664572 0.668872 0.003891 \n",
|
|
"13 0.647433 0.644864 0.649585 0.004972 \n",
|
|
"16 0.646960 0.645436 0.649530 0.004754 \n",
|
|
"4 0.644690 0.644991 0.648035 0.004519 \n",
|
|
"1 0.644901 0.644425 0.647983 0.004699 \n",
|
|
"19 0.645586 0.643166 0.647959 0.005163 \n",
|
|
"22 0.645409 0.643514 0.647726 0.004680 \n",
|
|
"7 0.643691 0.642749 0.646389 0.004498 \n",
|
|
"10 0.642975 0.643421 0.646229 0.004290 \n",
|
|
"14 0.577411 0.580170 0.581560 0.004075 \n",
|
|
"17 0.577197 0.580763 0.581224 0.003492 \n",
|
|
"2 0.576987 0.579852 0.581062 0.003915 \n",
|
|
"5 0.576217 0.580320 0.580723 0.003855 \n",
|
|
"20 0.572833 0.576280 0.576736 0.003389 \n",
|
|
"8 0.572391 0.575731 0.576321 0.003475 \n",
|
|
"23 0.572373 0.576699 0.576280 0.003034 \n",
|
|
"11 0.571607 0.576352 0.575901 0.003337 \n",
|
|
"\n",
|
|
" rank_test_score \n",
|
|
"12 1 \n",
|
|
"15 2 \n",
|
|
"18 3 \n",
|
|
"21 4 \n",
|
|
"3 5 \n",
|
|
"0 6 \n",
|
|
"9 7 \n",
|
|
"6 8 \n",
|
|
"13 9 \n",
|
|
"16 10 \n",
|
|
"4 11 \n",
|
|
"1 12 \n",
|
|
"19 13 \n",
|
|
"22 14 \n",
|
|
"7 15 \n",
|
|
"10 16 \n",
|
|
"14 17 \n",
|
|
"17 18 \n",
|
|
"2 19 \n",
|
|
"5 20 \n",
|
|
"20 21 \n",
|
|
"8 22 \n",
|
|
"23 23 \n",
|
|
"11 24 "
|
|
]
|
|
},
|
|
"execution_count": 4,
|
|
"metadata": {},
|
|
"output_type": "execute_result"
|
|
}
|
|
],
|
|
"source": [
|
|
"from sklearn.model_selection import GridSearchCV\n",
|
|
"import pandas as pd\n",
|
|
"\n",
|
|
"optimisation_pipeline = GridSearchCV(\n",
|
|
" create_pipeline(),\n",
|
|
" {\n",
|
|
" \"vectorizer__min_df\": [5, 20, 100],\n",
|
|
" \"vectorizer__max_df\": [0.05, 0.1],\n",
|
|
" \"classifier__alpha\": [0.5, 1],\n",
|
|
" \"classifier__fit_prior\": [True, False],\n",
|
|
" },\n",
|
|
" scoring=\"f1_macro\",\n",
|
|
" cv=3,\n",
|
|
" n_jobs=-1,\n",
|
|
" verbose=1,\n",
|
|
")\n",
|
|
"optimisation_pipeline.fit(X_train, y_train)\n",
|
|
"\n",
|
|
"results = pd.DataFrame(optimisation_pipeline.cv_results_)\n",
|
|
"results.sort_values(\"rank_test_score\")"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 5,
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"data": {
|
|
"text/html": [
|
|
"<style>#sk-container-id-1 {color: black;background-color: white;}#sk-container-id-1 pre{padding: 0;}#sk-container-id-1 div.sk-toggleable {background-color: white;}#sk-container-id-1 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-1 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-1 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-1 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-1 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-1 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-1 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-1 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-1 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-1 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-1 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-1 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-1 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-1 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-1 div.sk-item {position: relative;z-index: 1;}#sk-container-id-1 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-1 div.sk-item::before, #sk-container-id-1 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-1 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-1 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-1 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-1 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-1 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-1 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-1 div.sk-label-container {text-align: center;}#sk-container-id-1 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-1 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-1\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>Pipeline(steps=[('vectorizer',\n",
|
|
" TfidfVectorizer(max_df=0.05, min_df=5, sublinear_tf=True)),\n",
|
|
" ('classifier', MultinomialNB(alpha=1))])</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-1\" type=\"checkbox\" ><label for=\"sk-estimator-id-1\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">Pipeline</label><div class=\"sk-toggleable__content\"><pre>Pipeline(steps=[('vectorizer',\n",
|
|
" TfidfVectorizer(max_df=0.05, min_df=5, sublinear_tf=True)),\n",
|
|
" ('classifier', MultinomialNB(alpha=1))])</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-2\" type=\"checkbox\" ><label for=\"sk-estimator-id-2\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">TfidfVectorizer</label><div class=\"sk-toggleable__content\"><pre>TfidfVectorizer(max_df=0.05, min_df=5, sublinear_tf=True)</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-3\" type=\"checkbox\" ><label for=\"sk-estimator-id-3\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">MultinomialNB</label><div class=\"sk-toggleable__content\"><pre>MultinomialNB(alpha=1)</pre></div></div></div></div></div></div></div>"
|
|
],
|
|
"text/plain": [
|
|
"Pipeline(steps=[('vectorizer',\n",
|
|
" TfidfVectorizer(max_df=0.05, min_df=5, sublinear_tf=True)),\n",
|
|
" ('classifier', MultinomialNB(alpha=1))])"
|
|
]
|
|
},
|
|
"execution_count": 5,
|
|
"metadata": {},
|
|
"output_type": "execute_result"
|
|
}
|
|
],
|
|
"source": [
|
|
"from sklearn import set_config\n",
|
|
"\n",
|
|
"set_config(display=\"diagram\")\n",
|
|
"\n",
|
|
"classifier = create_pipeline()\n",
|
|
"classifier.set_params(**optimisation_pipeline.best_params_)\n",
|
|
"classifier.fit(X_train, y_train)"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 12,
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
" precision recall f1-score support\n",
|
|
"\n",
|
|
" business 0.6412 0.7258 0.6808 3198\n",
|
|
" economics 0.6635 0.7146 0.6881 3189\n",
|
|
" geography 0.7461 0.6791 0.7111 3207\n",
|
|
" medicine 0.8806 0.9021 0.8912 3187\n",
|
|
" politics 0.5563 0.5895 0.5724 3169\n",
|
|
" psychology 0.7654 0.6811 0.7208 3252\n",
|
|
" sociology 0.5165 0.4698 0.4921 3197\n",
|
|
"\n",
|
|
" accuracy 0.6803 22399\n",
|
|
" macro avg 0.6814 0.6803 0.6795 22399\n",
|
|
"weighted avg 0.6817 0.6803 0.6796 22399\n",
|
|
"\n"
|
|
]
|
|
},
|
|
{
|
|
"data": {
|
|
"image/png": "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",
|
|
"text/plain": [
|
|
"<Figure size 1440x1296 with 2 Axes>"
|
|
]
|
|
},
|
|
"metadata": {},
|
|
"output_type": "display_data"
|
|
}
|
|
],
|
|
"source": [
|
|
"from sklearn import metrics\n",
|
|
"import matplotlib.pyplot as plt\n",
|
|
"\n",
|
|
"y_predicted = classifier.predict(X_test)\n",
|
|
"\n",
|
|
"%matplotlib inline\n",
|
|
"plt.rcParams[\"figure.figsize\"] = (20, 18)\n",
|
|
"plt.rcParams[\"figure.facecolor\"] = \"white\"\n",
|
|
"plt.rcParams[\"font.size\"] = 22\n",
|
|
"plt.rcParams[\"axes.xmargin\"] = 0\n",
|
|
"\n",
|
|
"print(metrics.classification_report(y_test, y_predicted, digits=4))\n",
|
|
"metrics.ConfusionMatrixDisplay.from_predictions(\n",
|
|
" y_true=y_test,\n",
|
|
" y_pred=y_predicted,\n",
|
|
" xticks_rotation=\"vertical\",\n",
|
|
" normalize=\"pred\",\n",
|
|
" values_format=\".2f\",\n",
|
|
")\n",
|
|
"plt.tight_layout()\n",
|
|
"plt.savefig(\"mag-confusion.png\", dpi=600)"
|
|
]
|
|
}
|
|
],
|
|
"metadata": {
|
|
"kernelspec": {
|
|
"display_name": "Python 3.10.4 ('.env': venv)",
|
|
"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": "3.10.4"
|
|
},
|
|
"orig_nbformat": 4,
|
|
"vscode": {
|
|
"interpreter": {
|
|
"hash": "02dd6d3afbfa9fbbe1037d64ad9014965528a1ccad21929d6e72f466389a68ad"
|
|
}
|
|
}
|
|
},
|
|
"nbformat": 4,
|
|
"nbformat_minor": 2
|
|
}
|