Buckets:

hf-doc-build
/

doc

Files

xet

hf-doc-build/doc / transformers /main /ro /custom_tokenizers.html

HuggingFaceDocBuilder

about 15 hours ago

download

raw

29.2 kB

	<meta charset="utf-8" /><meta name="hf:doc:metadata" content="{"title":"Personalizarea tokenizerelor","local":"personalizarea-tokenizerelor","sections":[{"title":"Antrenarea unui tokenizer","local":"antrenarea-unui-tokenizer","sections":[],"depth":2},{"title":"Vocabular personalizat","local":"vocabular-personalizat","sections":[],"depth":2},{"title":"Subclasarea TokenizersBackend","local":"subclasarea-tokenizersbackend","sections":[],"depth":2}],"depth":1}"/>
	<link href="/docs/transformers/main/ro/_app/immutable/entry/start.BoS-cO03.js" rel="modulepreload">
	<link href="/docs/transformers/main/ro/_app/immutable/chunks/Y0KuiM1h.js" rel="modulepreload">
	<link href="/docs/transformers/main/ro/_app/immutable/chunks/Cl6kX2N0.js" rel="modulepreload">
	<link href="/docs/transformers/main/ro/_app/immutable/entry/app.4h203SCu.js" rel="modulepreload">
	<link href="/docs/transformers/main/ro/_app/immutable/chunks/BzzeShVL.js" rel="modulepreload">
	<link href="/docs/transformers/main/ro/_app/immutable/chunks/DWwSmaYb.js" rel="modulepreload">
	<link href="/docs/transformers/main/ro/_app/immutable/chunks/DM0RASQS.js" rel="modulepreload">
	<link href="/docs/transformers/main/ro/_app/immutable/chunks/DsnmJJEf.js" rel="modulepreload">
	<link href="/docs/transformers/main/ro/_app/immutable/nodes/0.uOquA-Sy.js" rel="modulepreload">
	<link href="/docs/transformers/main/ro/_app/immutable/chunks/C8aKV6he.js" rel="modulepreload">
	<link href="/docs/transformers/main/ro/_app/immutable/nodes/10.CEQsdp6u.js" rel="modulepreload">
	<link href="/docs/transformers/main/ro/_app/immutable/chunks/CjTxHwdC.js" rel="modulepreload">
	<!--hrcnld--><meta name="hf:doc:metadata" content="{"title":"Personalizarea tokenizerelor","local":"personalizarea-tokenizerelor","sections":[{"title":"Antrenarea unui tokenizer","local":"antrenarea-unui-tokenizer","sections":[],"depth":2},{"title":"Vocabular personalizat","local":"vocabular-personalizat","sections":[],"depth":2},{"title":"Subclasarea TokenizersBackend","local":"subclasarea-tokenizersbackend","sections":[],"depth":2}],"depth":1}"/><!---->
	<link href="/docs/transformers/main/ro/_app/immutable/assets/0.tn0RQdqM.css" rel="modulepreload"> <!--[--><!--[0--><!--[--><!--[0--><!--[--><p></p> <div class="items-center shrink-0 min-w-[100px] max-sm:min-w-[50px] justify-end ml-auto flex" style="float: right; margin-left: 10px; display: inline-flex; position: relative; z-index: 10;"><div class="inline-flex rounded-md max-sm:rounded-sm"><button class="inline-flex items-center gap-1 h-7 max-sm:h-7 px-2 max-sm:px-1.5 text-sm font-medium text-gray-800 border border-r-0 rounded-l-md max-sm:rounded-l-sm border-gray-200 bg-white hover:shadow-inner dark:border-gray-850 dark:bg-gray-950 dark:text-gray-200 dark:hover:bg-gray-800" aria-live="polite"><span class="inline-flex items-center justify-center rounded-md p-0.5 max-sm:p-0 hover:text-gray-800 dark:hover:text-gray-200"><svg class="sm:size-3.5 size-3" xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----></span> <span>Copy page</span></button> <button class="inline-flex items-center justify-center w-6 max-sm:w-5 h-7 max-sm:h-7 disabled:pointer-events-none text-sm text-gray-500 hover:text-gray-700 dark:hover:text-white rounded-r-md max-sm:rounded-r-sm border border-l transition border-gray-200 bg-white hover:shadow-inner dark:border-gray-850 dark:bg-gray-950 dark:text-gray-200 dark:hover:bg-gray-800" aria-haspopup="menu" aria-expanded="false" aria-label="Open copy menu"><svg class="transition-transform text-gray-400 overflow-visible sm:size-3.5 size-3 rotate-0" width="1em" height="1em" viewBox="0 0 12 7" fill="none" xmlns="http://www.w3.org/2000/svg"><path d="M1 1L6 6L11 1" stroke="currentColor"></path></svg><!----></button></div> <!--[-1--><!--]--></div><!----> <!--[0--><h1 class="relative group"><a id="personalizarea-tokenizerelor" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#personalizarea-tokenizerelor"><span><svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg><!----></span></a> <span>Personalizarea tokenizerelor</span></h1><!--]--><!----> <p>Tokenizerele sunt decuplate de vocabularele lor învățate. Asta îți permite să inițializezi un tokenizer gol pentru antrenare sau să creezi unul direct cu propriul vocabular. Pipeline-ul de bază pentru tokenization rămâne același (normalizer, pre-tokenizer, algoritmul de tokenization), deci nu trebuie să îl recreezi de la zero.</p> <p>Acest ghid îți arată cum să antrenezi și să creezi un tokenizer personalizat.</p> <!--[1--><h2 class="relative group"><a id="antrenarea-unui-tokenizer" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#antrenarea-unui-tokenizer"><span><svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg><!----></span></a> <span>Antrenarea unui tokenizer</span></h2><!--]--><!----> <p>Un tokenizer gol antrenabil înlocuiește vocabularul cu un nou vocabular țintă. Este util pentru adaptarea la un nou domeniu, cum ar fi finanțe, o limbă cu resurse reduse sau cod.</p> <p>Creează un tokenizer gol și încarcă un dataset.</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5 text-gray-600 " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Copied</div><!----></button><!----></div> <pre class="language-py "><!----><span class="hljs-keyword">from</span> datasets <span class="hljs-keyword">import</span> load_dataset
	<span class="hljs-keyword">from</span> transformers <span class="hljs-keyword">import</span> GemmaTokenizer

	tokenizer = GemmaTokenizer()
	dataset = load_dataset(<span class="hljs-string">"Josephgflowers/Finance-Instruct-500k"</span>, split=<span class="hljs-string">"train"</span>)<!----></pre></div><!----> <p>Folosește metoda <code>TokenizersBackend.train_new_from_iterator()</code> ca să antrenezi tokenizerul. Metoda acceptă o funcție generator ca să returneze bucăți de text din dataset în loc să încarce totul în memorie dintr-o dată. Argumentul <code>vocab_size</code> setează dimensiunea vocabularului tokenizer-ului.</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5 text-gray-600 " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Copied</div><!----></button><!----></div> <pre class="language-py "><!----><span class="hljs-keyword">def</span> <span class="hljs-title function_">batch_iterator</span>(<span class="hljs-params">batch_size=<span class="hljs-number">1000</span></span>):
	<span class="hljs-keyword">for</span> i <span class="hljs-keyword">in</span> <span class="hljs-built_in">range</span>(<span class="hljs-number">0</span>, <span class="hljs-built_in">len</span>(dataset), batch_size):
	<span class="hljs-keyword">yield</span> dataset[i : i + batch_size][<span class="hljs-string">"assistant"</span>]

	trained_tokenizer = tokenizer.train_new_from_iterator(
	batch_iterator(),
	vocab_size=<span class="hljs-number">32000</span>,
	)
	encoded = trained_tokenizer(<span class="hljs-string">"The stock market rallied today."</span>)
	<span class="hljs-built_in">print</span>(encoded[<span class="hljs-string">"input_ids"</span>])
	[<span class="hljs-number">5866</span>, <span class="hljs-number">11503</span>, <span class="hljs-number">98</span>, <span class="hljs-number">5885</span>, <span class="hljs-number">8617</span>, <span class="hljs-number">13381</span>, <span class="hljs-number">30</span>]<!----></pre></div><!----> <p>Adaugă token-uri speciale noi cu argumentul <code>new_special_tokens</code> sau folosește <code>special_tokens_map</code> ca să redenumești token-urile speciale vechi cu cele noi.</p> <p>Salvează noul tokenizer de finanțe cu <code>save_pretrained()</code> sau salvează-l și încarcă-l pe Hub cu <code>push_to_hub()</code>. Asta creează un fișier <code>tokenizer.json</code> care captează vocabularul nou antrenat, regulile de îmbinare și configurația completă a pipeline-ului.</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5 text-gray-600 " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Copied</div><!----></button><!----></div> <pre class="language-py "><!---->trained_tokenizer.save_pretrained(<span class="hljs-string">"./finance-gemma-tokenizer"</span>)
	trained_tokenizer.push_to_hub(<span class="hljs-string">"finance-gemma-tokenizer"</span>)<!----></pre></div><!----> <!--[1--><h2 class="relative group"><a id="vocabular-personalizat" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#vocabular-personalizat"><span><svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg><!----></span></a> <span>Vocabular personalizat</span></h2><!--]--><!----> <p>Un tokenizer gol suportă vocabular personalizat cu argumentele <code>vocab</code> și <code>merges</code>.</p> <ul><li><code>vocab</code> este setul complet de token-uri pe care un tokenizer le cunoaște, iar fiecare intrare mapează un token la input id-ul său.</li> <li><code>merges</code> definește cum ar trebui algoritmul BPE să combine token-urile adiacente.</li></ul> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5 text-gray-600 " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Copied</div><!----></button><!----></div> <pre class="language-py "><!----><span class="hljs-keyword">from</span> transformers <span class="hljs-keyword">import</span> GemmaTokenizer

	vocab={
	<span class="hljs-string">"<pad>"</span>: <span class="hljs-number">0</span>,
	<span class="hljs-string">"</s>"</span>: <span class="hljs-number">1</span>,
	<span class="hljs-string">"<s>"</span>: <span class="hljs-number">2</span>,
	<span class="hljs-string">"<unk>"</span>: <span class="hljs-number">3</span>,
	<span class="hljs-string">"<mask>"</span>: <span class="hljs-number">4</span>,
	<span class="hljs-string">"▁the"</span>: <span class="hljs-number">5</span>,
	<span class="hljs-string">"▁stock"</span>: <span class="hljs-number">6</span>,
	<span class="hljs-string">"▁market"</span>: <span class="hljs-number">7</span>,
	<span class="hljs-string">"▁"</span>: <span class="hljs-number">8</span>,
	<span class="hljs-string">"r"</span>: <span class="hljs-number">9</span>,
	<span class="hljs-string">"a"</span>: <span class="hljs-number">10</span>,
	<span class="hljs-string">"l"</span>: <span class="hljs-number">11</span>,
	<span class="hljs-string">"i"</span>: <span class="hljs-number">12</span>,
	<span class="hljs-string">"e"</span>: <span class="hljs-number">13</span>,
	<span class="hljs-string">"d"</span>: <span class="hljs-number">14</span>,
	<span class="hljs-string">"ra"</span>: <span class="hljs-number">15</span>,
	<span class="hljs-string">"li"</span>: <span class="hljs-number">16</span>,
	<span class="hljs-string">"lie"</span>: <span class="hljs-number">17</span>,
	<span class="hljs-string">"lied"</span>: <span class="hljs-number">18</span>,
	<span class="hljs-string">"ral"</span>: <span class="hljs-number">19</span>,
	<span class="hljs-string">"ralli"</span>: <span class="hljs-number">20</span>,
	<span class="hljs-string">"rallie"</span>: <span class="hljs-number">21</span>,
	<span class="hljs-string">"rallied"</span>: <span class="hljs-number">22</span>,
	}
	merges=[
	(<span class="hljs-string">"r"</span>, <span class="hljs-string">"a"</span>), <span class="hljs-comment"># r + a → ra</span>
	(<span class="hljs-string">"l"</span>, <span class="hljs-string">"i"</span>), <span class="hljs-comment"># l + i → li</span>
	(<span class="hljs-string">"li"</span>, <span class="hljs-string">"e"</span>), <span class="hljs-comment"># li + e → lie</span>
	(<span class="hljs-string">"lie"</span>, <span class="hljs-string">"d"</span>), <span class="hljs-comment"># lie + d → lied</span>
	(<span class="hljs-string">"ra"</span>, <span class="hljs-string">"l"</span>), <span class="hljs-comment"># ra + l → ral</span>
	(<span class="hljs-string">"ral"</span>, <span class="hljs-string">"li"</span>), <span class="hljs-comment"># ral + li → ralli</span>
	(<span class="hljs-string">"ralli"</span>, <span class="hljs-string">"e"</span>), <span class="hljs-comment"># ralli + e → rallie</span>
	(<span class="hljs-string">"rallie"</span>, <span class="hljs-string">"d"</span>), <span class="hljs-comment"># rallie + d → rallied</span>
	]

	tokenizer = GemmaTokenizer(vocab=vocab, merges=merges)
	encoded = tokenizer(<span class="hljs-string">"the stock market rallied"</span>)
	<span class="hljs-built_in">print</span>(encoded[<span class="hljs-string">"input_ids"</span>])<!----></pre></div><!----> <!--[1--><h2 class="relative group"><a id="subclasarea-tokenizersbackend" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#subclasarea-tokenizersbackend"><span><svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg><!----></span></a> <span>Subclasarea TokenizersBackend</span></h2><!--]--><!----> <p>Tokenizers suportă patru <a href="./fast_tokenizers#backend-uri">backend-uri</a> diferite. În general, ar trebui să folosești <code>TokenizersBackend</code> ca să definești un tokenizer nou deoarece este mai rapid.</p> <blockquote class="tip"><p><code>PythonBackend</code> este un tokenizer pur Python care nu depinde de backend-uri ca Rust, SentencePiece sau mistral-common. Folosește <code>PythonBackend</code> doar dacă construiești un tokenizer foarte specializat care nu poate fi exprimat de backend-ul Rust.</p></blockquote> <ol><li>Subclasează <code>TokenizersBackend</code> cu atribute de clasă precum latura de padding și algoritmul de tokenizare de folosit.</li> <li>Definește pipeline-ul de tokenizare în <code>__init__</code>. Asta include algoritmul de tokenizare de folosit, cum să împartă textul brut înaintea algoritmului și cum să decodifice token-urile înapoi în text.</li></ol> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5 text-gray-600 " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Copied</div><!----></button><!----></div> <pre class="language-py "><!----><span class="hljs-keyword">from</span> tokenizers <span class="hljs-keyword">import</span> Tokenizer, decoders, pre_tokenizers
	<span class="hljs-keyword">from</span> tokenizers.models <span class="hljs-keyword">import</span> BPE
	<span class="hljs-keyword">from</span> transformers <span class="hljs-keyword">import</span> TokenizersBackend

	<span class="hljs-keyword">class</span> <span class="hljs-title class_">NewTokenizer</span>(<span class="hljs-title class_ inherited__">TokenizersBackend</span>):
	padding_side = <span class="hljs-string">"left"</span>
	model = BPE

	<span class="hljs-keyword">def</span> <span class="hljs-title function_">__init__</span>(<span class="hljs-params">
	self,
	vocab=<span class="hljs-literal">None</span>,
	merges=<span class="hljs-literal">None</span>,
	unk_token=<span class="hljs-string">"<unk>"</span>,
	bos_token=<span class="hljs-string">"<s>"</span>,
	eos_token=<span class="hljs-string">"</s>"</span>,
	pad_token=<span class="hljs-string">"<pad>"</span>,
	</span>):
	<span class="hljs-variable language_">self</span>._vocab = vocab <span class="hljs-keyword">or</span> {
	<span class="hljs-built_in">str</span>(unk_token): <span class="hljs-number">0</span>,
	<span class="hljs-built_in">str</span>(bos_token): <span class="hljs-number">1</span>,
	<span class="hljs-built_in">str</span>(eos_token): <span class="hljs-number">2</span>,
	<span class="hljs-built_in">str</span>(pad_token): <span class="hljs-number">3</span>,
	}
	<span class="hljs-variable language_">self</span>._merges = merges <span class="hljs-keyword">or</span> []

	<span class="hljs-variable language_">self</span>._tokenizer = Tokenizer(
	BPE(vocab=<span class="hljs-variable language_">self</span>._vocab, merges=<span class="hljs-variable language_">self</span>._merges, fuse_unk=<span class="hljs-literal">True</span>)
	)
	<span class="hljs-variable language_">self</span>._tokenizer.pre_tokenizer = pre_tokenizers.ByteLevel(add_prefix_space=<span class="hljs-literal">False</span>)
	<span class="hljs-variable language_">self</span>._tokenizer.decoder = decoders.ByteLevel()

	<span class="hljs-built_in">super</span>().__init__(
	unk_token=unk_token,
	bos_token=bos_token,
	eos_token=eos_token,
	pad_token=pad_token,
	)<!----></pre></div><!----> <p>Antrenează sau salvează noul tokenizer gol.</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5 text-gray-600 " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Copied</div><!----></button><!----></div> <pre class="language-py "><!---->tokenizer = NewTokenizer()

	<span class="hljs-comment"># antrenează pe corpus nou</span>
	tokenizer.train_new_from_iterator()
	<span class="hljs-comment"># salvează tokenizer-ul</span>
	tokenizer.save_pretrained(<span class="hljs-string">"./new-tokenizer"</span>)<!----></pre></div><!----> <a class="!text-gray-400 !no-underline text-sm flex items-center not-prose mt-4" href="https://github.com/huggingface/transformers/blob/main/docs/source/ro/custom_tokenizers.md" target="_blank"><svg class="mr-1" xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M31,16l-7,7l-1.41-1.41L28.17,16l-5.58-5.59L24,9l7,7z"></path><path d="M1,16l7-7l1.41,1.41L3.83,16l5.58,5.59L8,23l-7-7z"></path><path d="M12.419,25.484L17.639,6.552l1.932,0.518L14.351,26.002z"></path></svg><!----> <span><span class="underline">Update</span> on GitHub</span></a><!----> <p></p><!--]--><!----><!--]--><!--]--><!--]--> <!--[-1--><!--]--><!--]-->

	<script>
	{
	__sveltekit_gltjkh = {
	base: "/docs/transformers/main/ro",
	assets: "/docs/transformers/main/ro"
	};

	const element = document.currentScript.parentElement;

	Promise.all([
	import("/docs/transformers/main/ro/_app/immutable/entry/start.BoS-cO03.js"),
	import("/docs/transformers/main/ro/_app/immutable/entry/app.4h203SCu.js")
	]).then(([kit, app]) => {
	kit.start(app, element, {
	node_ids: [0, 10],
	data: [null,null],
	form: null,
	error: null
	});
	});
	}
	</script>

Xet Storage Details

Size:: 29.2 kB
Xet hash:: e4481363cef9817bf0b297b53aa1db8670abf8f5d2b784dfd1f587ec05fb2353

Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.