Skip to content
Navigation Menu
{{ message }}
-
Notifications
You must be signed in to change notification settings - Fork 4
Expand file tree
/
Copy pathnlp_data.html
More file actions
399 lines (285 loc) · 21 KB
/
Copy pathnlp_data.html
File metadata and controls
399 lines (285 loc) · 21 KB
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
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
---
title: NLP Data
keywords: fastai
sidebar: home_sidebar
summary: "Handling NLP data"
description: "Handling NLP data"
nb_path: "nbs\41_nlp_data.ipynb"
---
<!--
#################################################
### THIS FILE WAS AUTOGENERATED! DO NOT EDIT! ###
#################################################
# file to edit: nbs\41_nlp_data.ipynb
# command to build the docs after a change: nbdev_build_docs
-->
<div class="container" id="notebook-container">
{% raw %}
<div class="cell border-box-sizing code_cell rendered">
</div>
{% endraw %}
<div class="cell border-box-sizing text_cell rendered"><div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<h2 id="Textual-data">Textual data<a class="anchor-link" href="#Textual-data"> </a></h2><blockquote><p>Raw & pure textual data</p>
</blockquote>
</div>
</div>
</div>
{% raw %}
<div class="cell border-box-sizing code_cell rendered">
</div>
{% endraw %}
<div class="cell border-box-sizing text_cell rendered"><div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<p><a href="/unpackai/nlp_data.html#Textual"><code>Textual</code></a> can help you</p>
<h3 id="Obtaining-data">Obtaining data<a class="anchor-link" href="#Obtaining-data"> </a></h3><p>We can extract data from following sources:</p>
<ul>
<li>a URL (http or https)</li>
<li>a path to a textual file (<em>.txt</em>, not <em>.doc</em>/<em>.docx</em>)</li>
<li>an interactive interface in the notebook, asking to upload the file</li>
</ul>
<h3 id="Prepare-model-training">Prepare model training<a class="anchor-link" href="#Prepare-model-training"> </a></h3><p>We can prepare model training in the following way:</p>
<ul>
<li>Pytorch datasets, data collector for dataloader</li>
<li>Create a hugging face trainer</li>
</ul>
</div>
</div>
</div>
{% raw %}
<div class="cell border-box-sizing code_cell rendered">
</div>
{% endraw %}
{% raw %}
<div class="cell border-box-sizing code_cell rendered">
<div class="output_wrapper">
<div class="output">
<div class="output_area">
<div class="output_markdown rendered_html output_subarea ">
<h4 id="Textual.create_datasets" class="doc_header"><code>Textual.create_datasets</code><a href="https://github.com/unpackai/unpackai/tree/master/unpackai/nlp/data.py#L114" class="source_link" style="float:right">[source]</a></h4><blockquote><p><code>Textual.create_datasets</code>(<strong><code>tokenizer</code></strong>, <strong><code>valid_ratio</code></strong>:<code>float</code>=<em><code>0.2</code></em>, <strong><code>train_path</code></strong>:<code>str</code>=<em><code>'./train_text.txt'</code></em>, <strong><code>val_path</code></strong>:<code>str</code>=<em><code>'./val_text.txt'</code></em>, <strong><code>block_size</code></strong>:<code>int</code>=<em><code>128</code></em>, <strong><code>mlm</code></strong>:<code>bool</code>=<em><code>False</code></em>)</p>
</blockquote>
<p>Create pytorch datasets and collating functions</p>
<ul>
<li>tokenizer: a huggingface tokenizer</li>
<li>valid ratio: portion of the valid data,
compare to the entire dataset</li>
<li>train_path: a path saving train text file</li>
<li>val_path: a path saving valid text file</li>
<li>block_size: max possible length of the sequence</li>
<li>mlm, return a masked language modeling collating
default False</li>
</ul>
</div>
</div>
<div class="output_area">
<div class="output_markdown rendered_html output_subarea ">
<h4 id="Textual.create_train_val" class="doc_header"><code>Textual.create_train_val</code><a href="https://github.com/unpackai/unpackai/tree/master/unpackai/nlp/data.py#L88" class="source_link" style="float:right">[source]</a></h4><blockquote><p><code>Textual.create_train_val</code>(<strong><code>valid_ratio</code></strong>=<em><code>0.2</code></em>, <strong><code>train_path</code></strong>=<em><code>'./train_text.txt'</code></em>, <strong><code>val_path</code></strong>=<em><code>'./val_text.txt'</code></em>)</p>
</blockquote>
<p>create 2 files:</p>
<ul>
<li>./train_text.txt</li>
<li>./val_text.txt</li>
</ul>
</div>
</div>
<div class="output_area">
<div class="output_markdown rendered_html output_subarea ">
<h4 id="Textual.from_path" class="doc_header"><code>Textual.from_path</code><a href="https://github.com/unpackai/unpackai/tree/master/unpackai/nlp/data.py#L52" class="source_link" style="float:right">[source]</a></h4><blockquote><p><code>Textual.from_path</code>(<strong><code>path</code></strong>:<code>Union</code>[<code>Path</code>, <code>str</code>])</p>
</blockquote>
<p>Load a textual object from system path</p>
</div>
</div>
<div class="output_area">
<div class="output_markdown rendered_html output_subarea ">
<h4 id="Textual.from_upload" class="doc_header"><code>Textual.from_upload</code><a href="https://github.com/unpackai/unpackai/tree/master/unpackai/nlp/data.py#L62" class="source_link" style="float:right">[source]</a></h4><blockquote><p><code>Textual.from_upload</code>(<strong><code>path</code></strong>:<code>Path</code>=<em><code>Path('uploaded_file.txt')</code></em>)</p>
</blockquote>
<p>Load textual with interactive upload button</p>
</div>
</div>
<div class="output_area">
<div class="output_markdown rendered_html output_subarea ">
<h4 id="Textual.from_url" class="doc_header"><code>Textual.from_url</code><a href="https://github.com/unpackai/unpackai/tree/master/unpackai/nlp/data.py#L47" class="source_link" style="float:right">[source]</a></h4><blockquote><p><code>Textual.from_url</code>(<strong><code>url</code></strong>:<code>str</code>)</p>
</blockquote>
<p>Create a Textual object from a URL</p>
</div>
</div>
<div class="output_area">
<div class="output_markdown rendered_html output_subarea ">
<h4 id="Textual.get_hf_trainer" class="doc_header"><code>Textual.get_hf_trainer</code><a href="https://github.com/unpackai/unpackai/tree/master/unpackai/nlp/data.py#L161" class="source_link" style="float:right">[source]</a></h4><blockquote><p><code>Textual.get_hf_trainer</code>(<strong><code>model</code></strong>, <strong><code>tokenizer</code></strong>, <strong><code>arguments</code></strong>=<em><code>None</code></em>, <strong><code>valid_ratio</code></strong>:<code>float</code>=<em><code>0.2</code></em>, <strong><code>train_path</code></strong>:<code>str</code>=<em><code>'./train_text.txt'</code></em>, <strong><code>val_path</code></strong>:<code>str</code>=<em><code>'./val_text.txt'</code></em>, <strong><code>block_size</code></strong>:<code>int</code>=<em><code>128</code></em>, <strong><code>mlm</code></strong>:<code>bool</code>=<em><code>False</code></em>)</p>
</blockquote>
<p>Create a hugging face trainer</p>
</div>
</div>
<div class="output_area">
<div class="output_markdown rendered_html output_subarea ">
<h4 id="Textual.show_batch" class="doc_header"><code>Textual.show_batch</code><a href="https://github.com/unpackai/unpackai/tree/master/unpackai/nlp/data.py#L105" class="source_link" style="float:right">[source]</a></h4><blockquote><p><code>Textual.show_batch</code>(<strong><code>tokenizer</code></strong>, <strong><code>bs</code></strong>:<code>int</code>=<em><code>4</code></em>)</p>
</blockquote>
</div>
</div>
</div>
</div>
</div>
{% endraw %}
<div class="cell border-box-sizing text_cell rendered"><div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<h2 id="Examples">Examples<a class="anchor-link" href="#Examples"> </a></h2>
</div>
</div>
</div>
<div class="cell border-box-sizing text_cell rendered"><div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<h3 id="Create-interactive-uploading-widget">Create interactive uploading widget<a class="anchor-link" href="#Create-interactive-uploading-widget"> </a></h3>
</div>
</div>
</div>
{% raw %}
<div class="cell border-box-sizing code_cell rendered">
<div class="input">
<div class="inner_cell">
<div class="input_area">
<div class=" highlight hl-ipython3"><pre><span></span><span class="n">uploaded</span> <span class="o">=</span> <span class="n">Textual</span><span class="o">.</span><span class="n">from_upload</span><span class="p">()</span>
</pre></div>
</div>
</div>
</div>
</div>
{% endraw %}
{% raw %}
<div class="cell border-box-sizing code_cell rendered">
<div class="input">
<div class="inner_cell">
<div class="input_area">
<div class=" highlight hl-ipython3"><pre><span></span><span class="n">textual</span> <span class="o">=</span> <span class="n">uploaded</span><span class="p">()</span>
<span class="n">textual</span>
</pre></div>
</div>
</div>
</div>
</div>
{% endraw %}
<div class="cell border-box-sizing text_cell rendered"><div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<p>Preview textual data by page</p>
</div>
</div>
</div>
{% raw %}
<div class="cell border-box-sizing code_cell rendered">
<div class="input">
<div class="inner_cell">
<div class="input_area">
<div class=" highlight hl-ipython3"><pre><span></span><span class="n">textual</span><span class="p">()</span>
</pre></div>
</div>
</div>
</div>
</div>
{% endraw %}
<div class="cell border-box-sizing text_cell rendered"><div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<h2 id="HFTextBlock">HFTextBlock<a class="anchor-link" href="#HFTextBlock"> </a></h2>
</div>
</div>
</div>
<div class="cell border-box-sizing text_cell rendered"><div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<p>A <code>TransformBlock</code>, designed for huggingface tokenizers</p>
</div>
</div>
</div>
{% raw %}
<div class="cell border-box-sizing code_cell rendered">
<div class="output_wrapper">
<div class="output">
<div class="output_area">
<div class="output_markdown rendered_html output_subarea ">
<h4 id="HFTextBlock" class="doc_header"><code>HFTextBlock</code><a href="https://github.com/unpackai/unpackai/tree/master/unpackai/nlp/data.py#L198" class="source_link" style="float:right">[source]</a></h4><blockquote><p><code>HFTextBlock</code>(<strong><code>tokenizer</code></strong>, <strong>**<code>tk_kwargs</code></strong>)</p>
</blockquote>
<p>create Huggingface specialized fastai Block
tokenizer: PreTrainedTokenizer from huggingface
tk_kwargs: keyward arguments for tokenizer's <code>__call__</code></p>
</div>
</div>
</div>
</div>
</div>
{% endraw %}
{% raw %}
<div class="cell border-box-sizing code_cell rendered">
</div>
{% endraw %}
<div class="cell border-box-sizing text_cell rendered"><div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<h1 id="Tests">Tests<a class="anchor-link" href="#Tests"> </a></h1>
</div>
</div>
</div>
{% raw %}
<div class="cell border-box-sizing code_cell rendered">
<div class="input">
<div class="inner_cell">
<div class="input_area">
<div class=" highlight hl-ipython3"><pre><span></span><span class="c1"># For Test Cases (might have duplicate import because it will be in a dedicated file)</span>
<span class="kn">from</span> <span class="nn">pathlib</span> <span class="kn">import</span> <span class="n">Path</span>
<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">List</span>
<span class="kn">import</span> <span class="nn">pytest</span>
<span class="kn">from</span> <span class="nn">test_common.utils_4_tests</span> <span class="kn">import</span> <span class="n">DATA_DIR</span>
<span class="kn">from</span> <span class="nn">test_utils</span> <span class="kn">import</span> <span class="n">GITHUB_TEST_DATA_URL</span><span class="p">,</span> <span class="n">check_connection_github</span>
</pre></div>
</div>
</div>
</div>
</div>
{% endraw %}
{% raw %}
<div class="cell border-box-sizing code_cell rendered">
<div class="input">
<div class="inner_cell">
<div class="input_area">
<div class=" highlight hl-ipython3"><pre><span></span><span class="n">LOCAL_TEST_TXT</span> <span class="o">=</span> <span class="n">DATA_DIR</span> <span class="o">/</span> <span class="s2">"to_download.txt"</span>
<span class="n">GITHUB_TEST_TXT</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">"</span><span class="si">{</span><span class="n">GITHUB_TEST_DATA_URL</span><span class="si">}</span><span class="s2">/to_download.txt"</span>
<span class="n">LOCAL_TEST_TXT_UTF8</span> <span class="o">=</span> <span class="n">DATA_DIR</span> <span class="o">/</span> <span class="s2">"Deep learning.txt"</span>
<span class="n">GITHUB_TEST_TXT_UTF8</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">"</span><span class="si">{</span><span class="n">GITHUB_TEST_DATA_URL</span><span class="si">}</span><span class="s2">/Deep%20learning.txt"</span>
<span class="nd">@pytest</span><span class="o">.</span><span class="n">fixture</span><span class="p">(</span><span class="n">scope</span><span class="o">=</span><span class="s2">"session"</span><span class="p">)</span>
<span class="k">def</span> <span class="nf">test_txt_content</span><span class="p">():</span>
<span class="k">return</span> <span class="n">LOCAL_TEST_TXT</span><span class="o">.</span><span class="n">read_text</span><span class="p">()</span>
<span class="nd">@pytest</span><span class="o">.</span><span class="n">fixture</span><span class="p">(</span><span class="n">scope</span><span class="o">=</span><span class="s2">"session"</span><span class="p">)</span>
<span class="k">def</span> <span class="nf">local_textual</span><span class="p">(</span><span class="n">test_txt_content</span><span class="p">):</span>
<span class="k">return</span> <span class="n">Textual</span><span class="p">(</span><span class="n">test_txt_content</span><span class="p">)</span>
<span class="k">class</span> <span class="nc">Test_Textual</span><span class="p">:</span>
<span class="k">def</span> <span class="nf">test_init</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">local_textual</span><span class="p">,</span> <span class="n">test_txt_content</span><span class="p">):</span>
<span class="sd">"""Test initialization of Textual from text"""</span>
<span class="n">expected_txt</span> <span class="o">=</span> <span class="n">test_txt_content</span><span class="o">.</span><span class="n">replace</span><span class="p">(</span><span class="s2">"</span><span class="se">\n</span><span class="s2">"</span><span class="p">,</span> <span class="s2">" "</span><span class="p">)</span><span class="o">.</span><span class="n">replace</span><span class="p">(</span><span class="s2">"</span><span class="se">\r</span><span class="s2">"</span><span class="p">,</span> <span class="s2">""</span><span class="p">)</span>
<span class="k">assert</span> <span class="n">local_textual</span><span class="o">.</span><span class="n">text</span> <span class="o">==</span> <span class="n">expected_txt</span>
<span class="k">def</span> <span class="nf">test_init_encoding</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
<span class="sd">"""Test initialization of Textual from text"""</span>
<span class="n">content</span> <span class="o">=</span> <span class="n">LOCAL_TEST_TXT_UTF8</span><span class="o">.</span><span class="n">read_text</span><span class="p">(</span><span class="n">encoding</span><span class="o">=</span><span class="s2">"utf-8"</span><span class="p">)</span>
<span class="n">textual</span> <span class="o">=</span> <span class="n">Textual</span><span class="p">(</span><span class="n">content</span><span class="p">)</span>
<span class="k">assert</span> <span class="n">textual</span><span class="o">.</span><span class="n">text</span> <span class="o">==</span> <span class="n">content</span><span class="o">.</span><span class="n">replace</span><span class="p">(</span><span class="s2">"</span><span class="se">\n</span><span class="s2">"</span><span class="p">,</span> <span class="s2">" "</span><span class="p">)</span><span class="o">.</span><span class="n">replace</span><span class="p">(</span><span class="s2">"</span><span class="se">\r</span><span class="s2">"</span><span class="p">,</span> <span class="s2">""</span><span class="p">)</span>
<span class="k">def</span> <span class="nf">test_from_path</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">local_textual</span><span class="p">):</span>
<span class="sd">"""Test create Textual from path (existing)"""</span>
<span class="n">textual</span> <span class="o">=</span> <span class="n">Textual</span><span class="o">.</span><span class="n">from_path</span><span class="p">(</span><span class="n">LOCAL_TEST_TXT</span><span class="p">)</span>
<span class="k">assert</span> <span class="n">textual</span><span class="o">.</span><span class="n">text</span> <span class="o">==</span> <span class="n">local_textual</span><span class="o">.</span><span class="n">text</span>
<span class="k">def</span> <span class="nf">test_from_path_error</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
<span class="sd">"""Test extract Textual of file that does not exist"""</span>
<span class="k">with</span> <span class="n">pytest</span><span class="o">.</span><span class="n">raises</span><span class="p">(</span><span class="ne">FileExistsError</span><span class="p">):</span>
<span class="n">textual</span> <span class="o">=</span> <span class="n">Textual</span><span class="o">.</span><span class="n">from_path</span><span class="p">(</span><span class="s2">"does_not_exist.txt"</span><span class="p">)</span>
<span class="nd">@pytest</span><span class="o">.</span><span class="n">mark</span><span class="o">.</span><span class="n">github</span>
<span class="k">def</span> <span class="nf">test_from_url</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">check_connection_github</span><span class="p">,</span> <span class="n">local_textual</span><span class="p">):</span>
<span class="sd">"""Test extract Textual from URL"""</span>
<span class="n">textual</span> <span class="o">=</span> <span class="n">Textual</span><span class="o">.</span><span class="n">from_url</span><span class="p">(</span><span class="n">GITHUB_TEST_TXT</span><span class="p">)</span>
<span class="k">assert</span> <span class="n">textual</span><span class="o">.</span><span class="n">text</span> <span class="o">==</span> <span class="n">local_textual</span><span class="o">.</span><span class="n">text</span><span class="p">,</span> <span class="sa">f</span><span class="s2">"URL text: </span><span class="si">{</span><span class="n">textual</span><span class="o">.</span><span class="n">text</span><span class="si">}</span><span class="s2">"</span>
<span class="nd">@pytest</span><span class="o">.</span><span class="n">mark</span><span class="o">.</span><span class="n">github</span>
<span class="k">def</span> <span class="nf">test_from_url_non_ascii</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">check_connection_github</span><span class="p">):</span>
<span class="sd">"""Test extract Textual from URL with non-ascii characters"""</span>
<span class="n">textual</span> <span class="o">=</span> <span class="n">Textual</span><span class="o">.</span><span class="n">from_url</span><span class="p">(</span><span class="n">GITHUB_TEST_TXT_UTF8</span><span class="p">)</span>
<span class="n">content</span> <span class="o">=</span> <span class="n">LOCAL_TEST_TXT_UTF8</span><span class="o">.</span><span class="n">read_text</span><span class="p">(</span><span class="n">encoding</span><span class="o">=</span><span class="s2">"utf-8"</span><span class="p">)</span>
<span class="n">content</span> <span class="o">=</span> <span class="n">content</span><span class="o">.</span><span class="n">replace</span><span class="p">(</span><span class="s2">"</span><span class="se">\n</span><span class="s2">"</span><span class="p">,</span> <span class="s2">" "</span><span class="p">)</span><span class="o">.</span><span class="n">replace</span><span class="p">(</span><span class="s2">"</span><span class="se">\r</span><span class="s2">"</span><span class="p">,</span> <span class="s2">""</span><span class="p">)</span>
<span class="k">assert</span> <span class="n">textual</span><span class="o">.</span><span class="n">text</span> <span class="o">==</span> <span class="n">content</span><span class="p">,</span> <span class="sa">f</span><span class="s2">"URL text: </span><span class="si">{</span><span class="n">textual</span><span class="o">.</span><span class="n">text</span><span class="si">}</span><span class="s2">"</span>
</pre></div>
</div>
</div>
</div>
</div>
{% endraw %}
</div>
You can’t perform that action at this time.