import{s as go,o as uo,n as mt}from"../chunks/scheduler.bdbef820.js";import{S as fo,i as ho,g as a,s as n,r as u,A as _o,h as r,f as i,c as o,j as J,u as f,x as m,k as C,y as t,a as p,v as h,d as _,t as b,w as y}from"../chunks/index.33f81d56.js";import{T as Gn}from"../chunks/Tip.34194030.js";import{D as B}from"../chunks/Docstring.8b71f9f8.js";import{C as Cn}from"../chunks/CodeBlock.3bad7fc9.js";import{E as Jn}from"../chunks/ExampleCodeBlock.030724b7.js";import{H as gt,E as bo}from"../chunks/getInferenceSnippets.d10958e6.js";function yo(Z){let l,k=`A large number of these flags control the logits or the stopping criteria of the generation. Make sure you check the generate-related classes for a full description of the possible manipulations, as well as examples of their usage.`;return{c(){l=a("p"),l.innerHTML=k},l(g){l=r(g,"P",{"data-svelte-h":!0}),m(l)!=="svelte-1lhagsi"&&(l.innerHTML=k)},m(g,d){p(g,l,d)},p:mt,d(g){g&&i(l)}}}function Mo(Z){let l,k="Examples:",g,d,M;return d=new Cn({props:{code:"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",highlighted:`>>> from transformers import GenerationConfig >>> # Download configuration from huggingface.co and cache. >>> generation_config = GenerationConfig.from_pretrained("openai-community/gpt2") >>> # E.g. config was saved using *save_pretrained('./test/saved_model/')* >>> generation_config.save_pretrained("./test/saved_model/") >>> generation_config = GenerationConfig.from_pretrained("./test/saved_model/") >>> # You can also specify configuration names to your generation configuration file >>> generation_config.save_pretrained("./test/saved_model/", config_file_name="my_configuration.json") >>> generation_config = GenerationConfig.from_pretrained("./test/saved_model/", "my_configuration.json") >>> # If you'd like to try a minor variation to an existing configuration, you can also pass generation >>> # arguments to \`.from_pretrained()\`. Be mindful that typos and unused arguments will be ignored >>> generation_config, unused_kwargs = GenerationConfig.from_pretrained( ... "openai-community/gpt2", top_k=1, foo=False, do_sample=True, return_unused_kwargs=True ... ) >>> generation_config.top_k 1 >>> unused_kwargs {'foo': False}`,wrap:!1}}),{c(){l=a("p"),l.textContent=k,g=n(),u(d.$$.fragment)},l(s){l=r(s,"P",{"data-svelte-h":!0}),m(l)!=="svelte-kvfsh7"&&(l.textContent=k),g=o(s),f(d.$$.fragment,s)},m(s,v){p(s,l,v),p(s,g,v),h(d,s,v),M=!0},p:mt,i(s){M||(_(d.$$.fragment,s),M=!0)},o(s){b(d.$$.fragment,s),M=!1},d(s){s&&(i(l),i(g)),y(d,s)}}}function vo(Z){let l,k=`Most generation-controlling parameters are set in generation_config which, if not passed, will be set to the model’s default generation configuration. You can override any generation_config by passing the corresponding parameters to generate(), e.g. .generate(inputs, num_beams=4, do_sample=True).`,g,d,M=`For an overview of generation strategies and code examples, check out the following guide.`;return{c(){l=a("p"),l.innerHTML=k,g=n(),d=a("p"),d.innerHTML=M},l(s){l=r(s,"P",{"data-svelte-h":!0}),m(l)!=="svelte-1c5u34l"&&(l.innerHTML=k),g=o(s),d=r(s,"P",{"data-svelte-h":!0}),m(d)!=="svelte-fvlq1g"&&(d.innerHTML=M)},m(s,v){p(s,l,v),p(s,g,v),p(s,d,v)},p:mt,d(s){s&&(i(l),i(g),i(d))}}}function To(Z){let l,k="Examples:",g,d,M;return d=new Cn({props:{code:"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",highlighted:`>>> from transformers import GPT2Tokenizer, AutoModelForCausalLM >>> import numpy as np >>> tokenizer = GPT2Tokenizer.from_pretrained("gpt2") >>> model = AutoModelForCausalLM.from_pretrained("openai-community/gpt2") >>> tokenizer.pad_token_id = tokenizer.eos_token_id >>> inputs = tokenizer(["Today is"], return_tensors="pt") >>> # Example 1: Print the scores for each token generated with Greedy Search >>> outputs = model.generate(**inputs, max_new_tokens=5, return_dict_in_generate=True, output_scores=True) >>> transition_scores = model.compute_transition_scores( ... outputs.sequences, outputs.scores, normalize_logits=True ... ) >>> # input_length is the length of the input prompt for decoder-only models, like the GPT family, and 1 for >>> # encoder-decoder models, like BART or T5. >>> input_length = 1 if model.config.is_encoder_decoder else inputs.input_ids.shape[1] >>> generated_tokens = outputs.sequences[:, input_length:] >>> for tok, score in zip(generated_tokens[0], transition_scores[0]): ... # | token | token string | log probability | probability ... print(f"| {tok:5d} | {tokenizer.decode(tok):8s} | {score.numpy():.3f} | {np.exp(score.numpy()):.2%}") | 262 | the | -1.414 | 24.33% | 1110 | day | -2.609 | 7.36% | 618 | when | -2.010 | 13.40% | 356 | we | -1.859 | 15.58% | 460 | can | -2.508 | 8.14% >>> # Example 2: Reconstruct the sequence scores from Beam Search >>> outputs = model.generate( ... **inputs, ... max_new_tokens=5, ... num_beams=4, ... num_return_sequences=4, ... return_dict_in_generate=True, ... output_scores=True, ... ) >>> transition_scores = model.compute_transition_scores( ... outputs.sequences, outputs.scores, outputs.beam_indices, normalize_logits=False ... ) >>> # If you sum the generated tokens' scores and apply the length penalty, you'll get the sequence scores. >>> # Tip 1: recomputing the scores is only guaranteed to match with \`normalize_logits=False\`. Depending on the >>> # use case, you might want to recompute it with \`normalize_logits=True\`. >>> # Tip 2: the output length does NOT include the input length >>> output_length = np.sum(transition_scores.numpy() < 0, axis=1) >>> length_penalty = model.generation_config.length_penalty >>> reconstructed_scores = transition_scores.sum(axis=1) / (output_length**length_penalty) >>> print(np.allclose(outputs.sequences_scores, reconstructed_scores)) True`,wrap:!1}}),{c(){l=a("p"),l.textContent=k,g=n(),u(d.$$.fragment)},l(s){l=r(s,"P",{"data-svelte-h":!0}),m(l)!=="svelte-kvfsh7"&&(l.textContent=k),g=o(s),f(d.$$.fragment,s)},m(s,v){p(s,l,v),p(s,g,v),h(d,s,v),M=!0},p:mt,i(s){M||(_(d.$$.fragment,s),M=!0)},o(s){b(d.$$.fragment,s),M=!1},d(s){s&&(i(l),i(g)),y(d,s)}}}function wo(Z){let l,k=`Most generation-controlling parameters are set in generation_config which, if not passed, will be set to the model’s default generation configuration. You can override any generation_config by passing the corresponding parameters to generate, e.g. .generate(inputs, num_beams=4, do_sample=True).`,g,d,M=`For an overview of generation strategies and code examples, check out the following guide.`;return{c(){l=a("p"),l.innerHTML=k,g=n(),d=a("p"),d.innerHTML=M},l(s){l=r(s,"P",{"data-svelte-h":!0}),m(l)!=="svelte-1pahvb2"&&(l.innerHTML=k),g=o(s),d=r(s,"P",{"data-svelte-h":!0}),m(d)!=="svelte-fvlq1g"&&(d.innerHTML=M)},m(s,v){p(s,l,v),p(s,g,v),p(s,d,v)},p:mt,d(s){s&&(i(l),i(g),i(d))}}}function xo(Z){let l,k="Examples:",g,d,M;return d=new Cn({props:{code:"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",highlighted:`>>> from transformers import GPT2Tokenizer, TFAutoModelForCausalLM >>> import numpy as np >>> tokenizer = GPT2Tokenizer.from_pretrained("openai-community/gpt2") >>> model = TFAutoModelForCausalLM.from_pretrained("openai-community/gpt2") >>> tokenizer.pad_token_id = tokenizer.eos_token_id >>> inputs = tokenizer(["Today is"], return_tensors="tf") >>> # Example 1: Print the scores for each token generated with Greedy Search >>> outputs = model.generate(**inputs, max_new_tokens=5, return_dict_in_generate=True, output_scores=True) >>> transition_scores = model.compute_transition_scores( ... outputs.sequences, outputs.scores, normalize_logits=True ... ) >>> # input_length is the length of the input prompt for decoder-only models, like the GPT family, and 1 for >>> # encoder-decoder models, like BART or T5. >>> input_length = 1 if model.config.is_encoder_decoder else inputs.input_ids.shape[1] >>> generated_tokens = outputs.sequences[:, input_length:] >>> for tok, score in zip(generated_tokens[0], transition_scores[0]): ... # | token | token string | logits | probability ... print(f"| {tok:5d} | {tokenizer.decode(tok):8s} | {score.numpy():.3f} | {np.exp(score.numpy()):.2%}") | 262 | the | -1.414 | 24.33% | 1110 | day | -2.609 | 7.36% | 618 | when | -2.010 | 13.40% | 356 | we | -1.859 | 15.58% | 460 | can | -2.508 | 8.14% >>> # Example 2: Reconstruct the sequence scores from Beam Search >>> outputs = model.generate( ... **inputs, ... max_new_tokens=5, ... num_beams=4, ... num_return_sequences=4, ... return_dict_in_generate=True, ... output_scores=True, ... ) >>> transition_scores = model.compute_transition_scores( ... outputs.sequences, outputs.scores, outputs.beam_indices, normalize_logits=False ... ) >>> # If you sum the generated tokens' scores and apply the length penalty, you'll get the sequence scores. >>> # Tip: recomputing the scores is only guaranteed to match with \`normalize_logits=False\`. Depending on the >>> # use case, you might want to recompute it with \`normalize_logits=True\`. >>> output_length = np.sum(transition_scores.numpy() < 0, axis=1) >>> length_penalty = model.generation_config.length_penalty >>> reconstructed_scores = np.sum(transition_scores, axis=1) / (output_length**length_penalty) >>> print(np.allclose(outputs.sequences_scores, reconstructed_scores)) True`,wrap:!1}}),{c(){l=a("p"),l.textContent=k,g=n(),u(d.$$.fragment)},l(s){l=r(s,"P",{"data-svelte-h":!0}),m(l)!=="svelte-kvfsh7"&&(l.textContent=k),g=o(s),f(d.$$.fragment,s)},m(s,v){p(s,l,v),p(s,g,v),h(d,s,v),M=!0},p:mt,i(s){M||(_(d.$$.fragment,s),M=!0)},o(s){b(d.$$.fragment,s),M=!1},d(s){s&&(i(l),i(g)),y(d,s)}}}function ko(Z){let l,k,g,d,M,s,v,Un="각 프레임워크에는 해당하는 GenerationMixin 클래스에서 구현된 텍스트 생성을 위한 generate 메소드가 있습니다:",ut,ne,Bn='
  • PyTorch에서는 generate()가 GenerationMixin에 구현되어 있습니다.
  • TensorFlow에서는 generate()가 TFGenerationMixin에 구현되어 있습니다.
  • Flax/JAX에서는 generate()가 FlaxGenerationMixin에 구현되어 있습니다.
    • ',ft,oe,Zn='사용하는 프레임워크에 상관없이, generate 메소드는 GenerationConfig 클래스 인스턴스로 매개변수화 할 수 있습니다. generate 메소드의 동작을 제어하는 모든 생성 매개변수 목록을 확인하려면 이 클래스를 참조하세요.',ht,se,$n='모델의 생성 설정을 어떻게 확인하고, 기본값이 무엇인지, 매개변수를 어떻게 임시로 변경하는지, 그리고 사용자 지정 생성 설정을 만들고 저장하는 방법을 배우려면 텍스트 생성 전략 가이드를 참조하세요. 이 가이드는 토큰 스트리밍과 같은 관련 기능을 사용하는 방법도 설명합니다.',_t,ae,bt,T,re,Bt,Ce,In=`Class that holds a configuration for a generation task. A generate call supports the following generation methods for text-decoder, text-to-text, speech-to-text, and vision-to-text models:`,Zt,Ue,Wn="
  • greedy decoding if num_beams=1 and do_sample=False
  • multinomial sampling if num_beams=1 and do_sample=True
  • beam-search decoding if num_beams>1 and do_sample=False
  • beam-search multinomial sampling if num_beams>1 and do_sample=True
  • diverse beam-search decoding if num_beams>1 and num_beam_groups>1
  • constrained beam-search decoding if constraints!=None or force_words_ids!=None
  • assisted decoding if assistant_model or prompt_lookup_num_tokens is passed to .generate()
    • ",$t,Be,zn='To learn more about decoding strategies refer to the text generation strategies guide.',It,S,Wt,z,ie,zt,Ze,Hn='Instantiate a GenerationConfig from a generation configuration file.',Ht,E,Vt,q,le,Ft,$e,Vn=`Instantiates a GenerationConfig from a PretrainedConfig. This function is useful to convert legacy PretrainedConfig objects, which may contain generation parameters, into a stand-alone GenerationConfig.`,Xt,D,ce,Rt,Ie,Fn=`Save a generation configuration object to the directory save_directory, so that it can be re-loaded using the from_pretrained() class method.`,Nt,Y,de,Lt,We,Xn=`Updates attributes of this class instance with attributes from kwargs if they match existing attributes, returning all the unused kwargs.`,St,H,me,Et,ze,Rn=`Validates the values of the attributes of the GenerationConfig instance. Raises exceptions in the presence of parameterization that can be detected as incorrect from the configuration instance alone.`,qt,He,Nn=`Note that some parameters not validated here are best validated at generate runtime, as they may depend on other inputs and/or the model, such as parameters related to the generation length.`,Dt,Q,pe,Yt,Ve,Ln='Returns the generation mode triggered by the GenerationConfig instance.',yt,I,ge,Qt,Fe,Sn=`Class that holds arguments for watermark generation and should be passed into GenerationConfig during generate. See this paper for more details on the arguments.`,At,Xe,En="Accepts the following keys:",Pt,Re,qn=`
  • greenlist_ratio (float): Used for watermarking. The ratio of “green” tokens used to the vocabulary size. Defaults to 0.25.
  • bias (float): Used with watermarking. The bias added to the selected “green” tokens’ logits. Defaults to 2.0.
  • hashing_key (int): Hashing key used for watermarking. Defaults to 15485863 (the millionth prime).
  • seeding_scheme (str): Algorithm to use for watermarking. Accepts values:
  • context_width(int): The context length of previous tokens to use in seeding. Higher context length makes watermarking more robust.
    • `,Mt,ue,vt,x,fe,Ot,Ne,Dn=`A class containing all functions for auto-regressive text generation, to be used as a mixin in model classes. Inheriting from this class causes the model to have special generation-related behavior, such as loading a GenerationConfig at initialization time or ensuring generate-related tests are run in transformers CI.`,Kt,Le,Yn=`A model class should inherit from GenerationMixin to enable calling methods like generate, or when it has defined a custom generate method that relies on GenerationMixin, directly or indirectly, which approximately shares the same interface to public methods like generate. Three examples:`,en,Se,Qn=`
  • LlamaForCausalLM should inherit from GenerationMixin to enable calling generate and other public methods in the mixin;
  • BlipForQuestionAnswering has a custom generate method that approximately shares the same interface as GenerationMixin.generate (it has a few extra arguments, and the same output). That function also calls GenerationMixin.generate indirectly, through an inner model. As such, BlipForQuestionAnswering should inherit from GenerationMixin to benefit from all generation-related automation in our codebase;
  • BarkModel has a custom generate method and one of its inner models calls GenerationMixin.generate. However, its generate does not share the same interface as GenerationMixin.generate. In this case, BarkModel should NOT inherit from GenerationMixin, as it breaks the generate interface.
    • `,tn,Ee,An='The class exposes generate(), which can be used for:',nn,qe,Pn="
  • greedy decoding if num_beams=1 and do_sample=False
  • multinomial sampling if num_beams=1 and do_sample=True
  • beam-search decoding if num_beams>1 and do_sample=False
  • beam-search multinomial sampling if num_beams>1 and do_sample=True
  • diverse beam-search decoding if num_beams>1 and num_beam_groups>1
  • constrained beam-search decoding if constraints!=None or force_words_ids!=None
  • assisted decoding if assistant_model or prompt_lookup_num_tokens is passed to .generate()
    • ",on,De,On='To learn more about decoding strategies refer to the text generation strategies guide.',sn,V,he,an,Ye,Kn="Generates sequences of token ids for models with a language modeling head.",rn,A,ln,F,_e,cn,Qe,eo=`Computes the transition scores of sequences given the generation scores (and beam indices, if beam search was used). This is a convenient method to quickly obtain the scores of the selected tokens at generation time.`,dn,P,Tt,be,wt,j,ye,mn,Ae,to='A class containing all of the functions supporting generation, to be used as a mixin in TFPreTrainedModel.',pn,Pe,no='The class exposes generate(), which can be used for:',gn,Oe,oo=`
  • greedy decoding by calling greedy_search() if num_beams=1 and do_sample=False
  • contrastive search by calling contrastive_search() if penalty_alpha>0 and top_k>1
  • multinomial sampling by calling sample() if num_beams=1 and do_sample=True
  • beam-search decoding by calling beam_search() if num_beams>1
    • `,un,Ke,so=`You do not need to call any of the above methods directly. Pass custom parameter values to ‘generate’ instead. To learn more about decoding strategies refer to the text generation strategies guide.`,fn,X,Me,hn,et,ao="Generates sequences of token ids for models with a language modeling head.",_n,O,bn,R,ve,yn,tt,ro=`Computes the transition scores of sequences given the generation scores (and beam indices, if beam search was used). This is a convenient method to quickly obtain the scores of the selected tokens at generation time.`,Mn,K,xt,Te,kt,U,we,vn,nt,io=`A class containing all functions for auto-regressive text generation, to be used as a mixin in FlaxPreTrainedModel.`,Tn,ot,lo='The class exposes generate(), which can be used for:',wn,st,co=`
  • greedy decoding by calling _greedy_search() if num_beams=1 and do_sample=False
  • multinomial sampling by calling _sample() if num_beams=1 and do_sample=True
  • beam-search decoding by calling _beam_search() if num_beams>1 and do_sample=False
    • `,xn,at,mo=`You do not need to call any of the above methods directly. Pass custom parameter values to ‘generate’ instead. To learn more about decoding strategies refer to the text generation strategies guide.`,kn,ee,xe,jn,rt,po="Generates sequences of token ids for models with a language modeling head.",jt,ke,Gt,pt,Jt;return M=new gt({props:{title:"생성",local:"generation",headingTag:"h1"}}),ae=new gt({props:{title:"GenerationConfig",local:"transformers.GenerationConfig ][ transformers.GenerationConfig",headingTag:"h2"}}),re=new B({props:{name:"class transformers.GenerationConfig",anchor:"transformers.GenerationConfig",parameters:[{name:"**kwargs",val:""}],source:"https://github.com/huggingface/transformers/blob/v4.56.2/src/transformers/generation/configuration_utils.py#L82",parameterGroups:[{title:"Parameters that control the length of the output",parametersDescription:[{anchor:"transformers.GenerationConfig.max_length",description:`max_length (int, optional, defaults to 20) — The maximum length the generated tokens can have. Corresponds to the length of the input prompt + max_new_tokens. Its effect is overridden by max_new_tokens, if also set.`,name:"max_length"},{anchor:"transformers.GenerationConfig.max_new_tokens",description:`max_new_tokens (int, optional) — The maximum numbers of tokens to generate, ignoring the number of tokens in the prompt.`,name:"max_new_tokens"},{anchor:"transformers.GenerationConfig.min_length",description:`min_length (int, optional, defaults to 0) — The minimum length of the sequence to be generated. Corresponds to the length of the input prompt + min_new_tokens. Its effect is overridden by min_new_tokens, if also set.`,name:"min_length"},{anchor:"transformers.GenerationConfig.min_new_tokens",description:`min_new_tokens (int, optional) — The minimum numbers of tokens to generate, ignoring the number of tokens in the prompt.`,name:"min_new_tokens"},{anchor:"transformers.GenerationConfig.early_stopping",description:`early_stopping (bool or str, optional, defaults to False) — Controls the stopping condition for beam-based methods, like beam-search. It accepts the following values: True, where the generation stops as soon as there are num_beams complete candidates; False, where an heuristic is applied and the generation stops when is it very unlikely to find better candidates; "never", where the beam search procedure only stops when there cannot be better candidates (canonical beam search algorithm).`,name:"early_stopping"},{anchor:"transformers.GenerationConfig.max_time",description:`max_time (float, optional) — The maximum amount of time you allow the computation to run for in seconds. generation will still finish the current pass after allocated time has been passed.`,name:"max_time"},{anchor:"transformers.GenerationConfig.stop_strings",description:`stop_strings (str or list[str], optional) — A string or a list of strings that should terminate generation if the model outputs them.`,name:"stop_strings"}]},{title:"Parameters that control the generation strategy used",parametersDescription:[{anchor:"transformers.GenerationConfig.do_sample",description:`do_sample (bool, optional, defaults to False) — Whether or not to use sampling ; use greedy decoding otherwise.`,name:"do_sample"},{anchor:"transformers.GenerationConfig.num_beams",description:`num_beams (int, optional, defaults to 1) — Number of beams for beam search. 1 means no beam search.`,name:"num_beams"},{anchor:"transformers.GenerationConfig.num_beam_groups",description:`num_beam_groups (int, optional, defaults to 1) — Number of groups to divide num_beams into in order to ensure diversity among different groups of beams. this paper for more details.`,name:"num_beam_groups"}]},{title:"Parameters that control the cache",parametersDescription:[{anchor:"transformers.GenerationConfig.use_cache",description:`use_cache (bool, optional, defaults to True) — Whether or not the model should use the past last key/values attentions (if applicable to the model) to speed up decoding.`,name:"use_cache"},{anchor:"transformers.GenerationConfig.cache_implementation",description:`cache_implementation (str, optional, default to None) — Name of the cache class that will be instantiated in generate, for faster decoding. Possible values are:

    If none is specified, we will use the default cache for the model (which is often DynamicCache). See our cache documentation for further information.`,name:"cache_implementation"},{anchor:"transformers.GenerationConfig.cache_config",description:`cache_config (dict, optional, default to None) — Arguments used in the key-value cache class can be passed in cache_config.`,name:"cache_config"},{anchor:"transformers.GenerationConfig.return_legacy_cache",description:`return_legacy_cache (bool, optional, default to True) — Whether to return the legacy or new format of the cache when DynamicCache is used by default.`,name:"return_legacy_cache"}]},{title:"Parameters for manipulation of the model output logits",parametersDescription:[{anchor:"transformers.GenerationConfig.temperature",description:`temperature (float, optional, defaults to 1.0) — The value used to module the next token probabilities. This value is set in a model’s generation_config.json file. If it isn’t set, the default value is 1.0`,name:"temperature"},{anchor:"transformers.GenerationConfig.top_k",description:`top_k (int, optional, defaults to 50) — The number of highest probability vocabulary tokens to keep for top-k-filtering. This value is set in a model’s generation_config.json file. If it isn’t set, the default value is 50.`,name:"top_k"},{anchor:"transformers.GenerationConfig.top_p",description:`top_p (float, optional, defaults to 1.0) — If set to float < 1, only the smallest set of most probable tokens with probabilities that add up to top_p or higher are kept for generation. This value is set in a model’s generation_config.json file. If it isn’t set, the default value is 1.0`,name:"top_p"},{anchor:"transformers.GenerationConfig.min_p",description:`min_p (float, optional) — Minimum token probability, which will be scaled by the probability of the most likely token. It must be a value between 0 and 1. Typical values are in the 0.01-0.2 range, comparably selective as setting top_p in the 0.99-0.8 range (use the opposite of normal top_p values).`,name:"min_p"},{anchor:"transformers.GenerationConfig.typical_p",description:`typical_p (float, optional, defaults to 1.0) — Local typicality measures how similar the conditional probability of predicting a target token next is to the expected conditional probability of predicting a random token next, given the partial text already generated. If set to float < 1, the smallest set of the most locally typical tokens with probabilities that add up to typical_p or higher are kept for generation. See this paper for more details.`,name:"typical_p"},{anchor:"transformers.GenerationConfig.epsilon_cutoff",description:`epsilon_cutoff (float, optional, defaults to 0.0) — If set to float strictly between 0 and 1, only tokens with a conditional probability greater than epsilon_cutoff will be sampled. In the paper, suggested values range from 3e-4 to 9e-4, depending on the size of the model. See Truncation Sampling as Language Model Desmoothing for more details.`,name:"epsilon_cutoff"},{anchor:"transformers.GenerationConfig.eta_cutoff",description:`eta_cutoff (float, optional, defaults to 0.0) — Eta sampling is a hybrid of locally typical sampling and epsilon sampling. If set to float strictly between 0 and 1, a token is only considered if it is greater than either eta_cutoff or sqrt(eta_cutoff) * exp(-entropy(softmax(next_token_logits))). The latter term is intuitively the expected next token probability, scaled by sqrt(eta_cutoff). In the paper, suggested values range from 3e-4 to 2e-3, depending on the size of the model. See Truncation Sampling as Language Model Desmoothing for more details.`,name:"eta_cutoff"},{anchor:"transformers.GenerationConfig.diversity_penalty",description:`diversity_penalty (float, optional, defaults to 0.0) — This value is subtracted from a beam’s score if it generates a token same as any beam from other group at a particular time. Note that diversity_penalty is only effective if group beam search is enabled.`,name:"diversity_penalty"},{anchor:"transformers.GenerationConfig.repetition_penalty",description:`repetition_penalty (float, optional, defaults to 1.0) — The parameter for repetition penalty. 1.0 means no penalty. See this paper for more details.`,name:"repetition_penalty"},{anchor:"transformers.GenerationConfig.encoder_repetition_penalty",description:`encoder_repetition_penalty (float, optional, defaults to 1.0) — The parameter for encoder_repetition_penalty. An exponential penalty on sequences that are not in the original input. 1.0 means no penalty.`,name:"encoder_repetition_penalty"},{anchor:"transformers.GenerationConfig.length_penalty",description:`length_penalty (float, optional, defaults to 1.0) — Exponential penalty to the length that is used with beam-based generation. It is applied as an exponent to the sequence length, which in turn is used to divide the score of the sequence. Since the score is the log likelihood of the sequence (i.e. negative), length_penalty > 0.0 promotes longer sequences, while length_penalty < 0.0 encourages shorter sequences.`,name:"length_penalty"},{anchor:"transformers.GenerationConfig.no_repeat_ngram_size",description:`no_repeat_ngram_size (int, optional, defaults to 0) — If set to int > 0, all ngrams of that size can only occur once.`,name:"no_repeat_ngram_size"},{anchor:"transformers.GenerationConfig.bad_words_ids",description:`bad_words_ids (list[list[int]], optional) — List of list of token ids that are not allowed to be generated. Check NoBadWordsLogitsProcessor for further documentation and examples.`,name:"bad_words_ids"},{anchor:"transformers.GenerationConfig.force_words_ids",description:`force_words_ids (list[list[int]] or list[list[list[int]]], optional) — List of token ids that must be generated. If given a list[list[int]], this is treated as a simple list of words that must be included, the opposite to bad_words_ids. If given list[list[list[int]]], this triggers a disjunctive constraint, where one can allow different forms of each word.`,name:"force_words_ids"},{anchor:"transformers.GenerationConfig.renormalize_logits",description:`renormalize_logits (bool, optional, defaults to False) — Whether to renormalize the logits after applying all the logits processors (including the custom ones). It’s highly recommended to set this flag to True as the search algorithms suppose the score logits are normalized but some logit processors break the normalization.`,name:"renormalize_logits"},{anchor:"transformers.GenerationConfig.constraints",description:`constraints (list[Constraint], optional) — Custom constraints that can be added to the generation to ensure that the output will contain the use of certain tokens as defined by Constraint objects, in the most sensible way possible.`,name:"constraints"},{anchor:"transformers.GenerationConfig.forced_bos_token_id",description:`forced_bos_token_id (int, optional, defaults to model.config.forced_bos_token_id) — The id of the token to force as the first generated token after the decoder_start_token_id. Useful for multilingual models like mBART where the first generated token needs to be the target language token.`,name:"forced_bos_token_id"},{anchor:"transformers.GenerationConfig.forced_eos_token_id",description:"forced_eos_token_id (int or list[int], *optional*, defaults to model.config.forced_eos_token_id) -- The id of the token to force as the last generated token when max_length` is reached. Optionally, use a\nlist to set multiple end-of-sequence tokens.",name:"forced_eos_token_id"},{anchor:"transformers.GenerationConfig.remove_invalid_values",description:`remove_invalid_values (bool, optional, defaults to model.config.remove_invalid_values) — Whether to remove possible nan and inf outputs of the model to prevent the generation method to crash. Note that using remove_invalid_values can slow down generation.`,name:"remove_invalid_values"},{anchor:"transformers.GenerationConfig.exponential_decay_length_penalty",description:`exponential_decay_length_penalty (tuple(int, float), optional) — This Tuple adds an exponentially increasing length penalty, after a certain amount of tokens have been generated. The tuple shall consist of: (start_index, decay_factor) where start_index indicates where penalty starts and decay_factor represents the factor of exponential decay`,name:"exponential_decay_length_penalty"},{anchor:"transformers.GenerationConfig.suppress_tokens",description:`suppress_tokens (list[int], optional) — A list of tokens that will be suppressed at generation. The SupressTokens logit processor will set their log probs to -inf so that they are not sampled.`,name:"suppress_tokens"},{anchor:"transformers.GenerationConfig.begin_suppress_tokens",description:`begin_suppress_tokens (list[int], optional) — A list of tokens that will be suppressed at the beginning of the generation. The SupressBeginTokens logit processor will set their log probs to -inf so that they are not sampled.`,name:"begin_suppress_tokens"},{anchor:"transformers.GenerationConfig.sequence_bias",description:`sequence_bias (dict[tuple[int], float], optional)) — Dictionary that maps a sequence of tokens to its bias term. Positive biases increase the odds of the sequence being selected, while negative biases do the opposite. Check SequenceBiasLogitsProcessor for further documentation and examples.`,name:"sequence_bias"},{anchor:"transformers.GenerationConfig.token_healing",description:`token_healing (bool, optional, defaults to False) — Heal tail tokens of prompts by replacing them with their appropriate extensions. This enhances the quality of completions for prompts affected by greedy tokenization bias.`,name:"token_healing"},{anchor:"transformers.GenerationConfig.guidance_scale",description:`guidance_scale (float, optional) — The guidance scale for classifier free guidance (CFG). CFG is enabled by setting guidance_scale > 1. Higher guidance scale encourages the model to generate samples that are more closely linked to the input prompt, usually at the expense of poorer quality.`,name:"guidance_scale"},{anchor:"transformers.GenerationConfig.watermarking_config",description:`watermarking_config (BaseWatermarkingConfig or dict, optional) — Arguments used to watermark the model outputs by adding a small bias to randomly selected set of “green” tokens. See the docs of SynthIDTextWatermarkingConfig and WatermarkingConfig for more details. If passed as Dict, it will be converted to a WatermarkingConfig internally.`,name:"watermarking_config"}]},{title:"Parameters that define the output variables of generate",parametersDescription:[{anchor:"transformers.GenerationConfig.num_return_sequences",description:`num_return_sequences (int, optional, defaults to 1) — The number of independently computed returned sequences for each element in the batch.`,name:"num_return_sequences"},{anchor:"transformers.GenerationConfig.output_attentions",description:`output_attentions (bool, optional, defaults to False) — Whether or not to return the attentions tensors of all attention layers. See attentions under returned tensors for more details.`,name:"output_attentions"},{anchor:"transformers.GenerationConfig.output_hidden_states",description:`output_hidden_states (bool, optional, defaults to False) — Whether or not to return the hidden states of all layers. See hidden_states under returned tensors for more details.`,name:"output_hidden_states"},{anchor:"transformers.GenerationConfig.output_scores",description:`output_scores (bool, optional, defaults to False) — Whether or not to return the prediction scores. See scores under returned tensors for more details.`,name:"output_scores"},{anchor:"transformers.GenerationConfig.output_logits",description:`output_logits (bool, optional) — Whether or not to return the unprocessed prediction logit scores. See logits under returned tensors for more details.`,name:"output_logits"},{anchor:"transformers.GenerationConfig.return_dict_in_generate",description:`return_dict_in_generate (bool, optional, defaults to False) — Whether or not to return a ModelOutput, as opposed to returning exclusively the generated sequence. This flag must be set to True to return the generation cache (when use_cache is True) or optional outputs (see flags starting with output_)`,name:"return_dict_in_generate"}]},{title:"Special tokens that can be used at generation time",parametersDescription:[{anchor:"transformers.GenerationConfig.pad_token_id",description:`pad_token_id (int, optional) — The id of the padding token.`,name:"pad_token_id"},{anchor:"transformers.GenerationConfig.bos_token_id",description:`bos_token_id (int, optional) — The id of the beginning-of-sequence token.`,name:"bos_token_id"},{anchor:"transformers.GenerationConfig.eos_token_id",description:`eos_token_id (Union[int, list[int]], optional) — The id of the end-of-sequence token. Optionally, use a list to set multiple end-of-sequence tokens.`,name:"eos_token_id"}]},{title:"Generation parameters exclusive to encoder-decoder models",parametersDescription:[{anchor:"transformers.GenerationConfig.encoder_no_repeat_ngram_size",description:`encoder_no_repeat_ngram_size (int, optional, defaults to 0) — If set to int > 0, all ngrams of that size that occur in the encoder_input_ids cannot occur in the decoder_input_ids.`,name:"encoder_no_repeat_ngram_size"},{anchor:"transformers.GenerationConfig.decoder_start_token_id",description:`decoder_start_token_id (int or list[int], optional) — If an encoder-decoder model starts decoding with a different token than bos, the id of that token or a list of length batch_size. Indicating a list enables different start ids for each element in the batch (e.g. multilingual models with different target languages in one batch)`,name:"decoder_start_token_id"}]},{title:"Generation parameters exclusive to assistant generation",parametersDescription:[{anchor:"transformers.GenerationConfig.is_assistant",description:`is_assistant (bool, optional, defaults to False) — Whether the model is an assistant (draft) model.`,name:"is_assistant"},{anchor:"transformers.GenerationConfig.num_assistant_tokens",description:`num_assistant_tokens (int, optional, defaults to 20) — Defines the number of speculative tokens that shall be generated by the assistant model before being checked by the target model at each iteration. Higher values for num_assistant_tokens make the generation more speculative : If the assistant model is performant larger speed-ups can be reached, if the assistant model requires lots of corrections, lower speed-ups are reached.`,name:"num_assistant_tokens"},{anchor:"transformers.GenerationConfig.num_assistant_tokens_schedule",description:`num_assistant_tokens_schedule (str, optional, defaults to "constant") — Defines the schedule at which max assistant tokens shall be changed during inference.

    `,name:"num_assistant_tokens_schedule"},{anchor:"transformers.GenerationConfig.assistant_confidence_threshold",description:`assistant_confidence_threshold (float, optional, defaults to 0.4) — The confidence threshold for the assistant model. If the assistant model’s confidence in its prediction for the current token is lower than this threshold, the assistant model stops the current token generation iteration, even if the number of speculative tokens (defined by num_assistant_tokens) is not yet reached. The assistant’s confidence threshold is adjusted throughout the speculative iterations to reduce the number of unnecessary draft and target forward passes, biased towards avoiding false negatives. assistant_confidence_threshold value is persistent over multiple generation calls with the same assistant model. It is an unsupervised version of the dynamic speculation lookahead from Dynamic Speculation Lookahead Accelerates Speculative Decoding of Large Language Models https://huggingface.co/papers/2405.04304.`,name:"assistant_confidence_threshold"},{anchor:"transformers.GenerationConfig.prompt_lookup_num_tokens",description:`prompt_lookup_num_tokens (int, optional) — The number of tokens to be output as candidate tokens.`,name:"prompt_lookup_num_tokens"},{anchor:"transformers.GenerationConfig.max_matching_ngram_size",description:`max_matching_ngram_size (int, optional) — The maximum ngram size to be considered for matching in the prompt. Default to 2 if not provided.`,name:"max_matching_ngram_size"},{anchor:"transformers.GenerationConfig.assistant_early_exit(int,",description:`assistant_early_exit(int, optional) — If set to a positive integer, early exit of the model will be used as an assistant. Can only be used with models that support early exit (i.e. models where logits from intermediate layers can be interpreted by the LM head).`,name:"assistant_early_exit(int,"},{anchor:"transformers.GenerationConfig.assistant_lookbehind(int,",description:`assistant_lookbehind(int, optional, defaults to 10) — If set to a positive integer, the re-encodeing process will additionally consider the last assistant_lookbehind assistant tokens to correctly align tokens. Can only be used with different tokenizers in speculative decoding. See this blog for more details.`,name:"assistant_lookbehind(int,"},{anchor:"transformers.GenerationConfig.target_lookbehind(int,",description:`target_lookbehind(int, optional, defaults to 10) — If set to a positive integer, the re-encodeing process will additionally consider the last target_lookbehind target tokens to correctly align tokens. Can only be used with different tokenizers in speculative decoding. See this blog for more details.`,name:"target_lookbehind(int,"}]},{title:"Parameters related to performances and compilation",parametersDescription:[{anchor:"transformers.GenerationConfig.compile_config",description:`compile_config (CompileConfig, optional) — If using a compilable cache, this controls how generate will compile the forward pass for faster inference.`,name:"compile_config"},{anchor:"transformers.GenerationConfig.disable_compile",description:`disable_compile (bool, optional) — Whether to disable the automatic compilation of the forward pass. Automatic compilation happens when specific criteria are met, including using a compilable cache. Please open an issue if you find the need to use this flag.`,name:"disable_compile"}]}]}}),S=new Gn({props:{$$slots:{default:[yo]},$$scope:{ctx:Z}}}),ie=new B({props:{name:"from_pretrained",anchor:"transformers.GenerationConfig.from_pretrained",parameters:[{name:"pretrained_model_name",val:": typing.Union[str, os.PathLike]"},{name:"config_file_name",val:": typing.Union[str, os.PathLike, NoneType] = None"},{name:"cache_dir",val:": typing.Union[str, os.PathLike, NoneType] = None"},{name:"force_download",val:": bool = False"},{name:"local_files_only",val:": bool = False"},{name:"token",val:": typing.Union[bool, str, NoneType] = None"},{name:"revision",val:": str = 'main'"},{name:"**kwargs",val:""}],parametersDescription:[{anchor:"transformers.GenerationConfig.from_pretrained.pretrained_model_name",description:`pretrained_model_name (str or os.PathLike) — This can be either:

    `,name:"pretrained_model_name"},{anchor:"transformers.GenerationConfig.from_pretrained.config_file_name",description:`config_file_name (str or os.PathLike, optional, defaults to "generation_config.json") — Name of the generation configuration JSON file to be loaded from pretrained_model_name.`,name:"config_file_name"},{anchor:"transformers.GenerationConfig.from_pretrained.cache_dir",description:`cache_dir (str or os.PathLike, optional) — Path to a directory in which a downloaded pretrained model configuration should be cached if the standard cache should not be used.`,name:"cache_dir"},{anchor:"transformers.GenerationConfig.from_pretrained.force_download",description:`force_download (bool, optional, defaults to False) — Whether or not to force to (re-)download the configuration files and override the cached versions if they exist.`,name:"force_download"},{anchor:"transformers.GenerationConfig.from_pretrained.resume_download",description:`resume_download — Deprecated and ignored. All downloads are now resumed by default when possible. Will be removed in v5 of Transformers.`,name:"resume_download"},{anchor:"transformers.GenerationConfig.from_pretrained.proxies",description:`proxies (dict[str, str], optional) — A dictionary of proxy servers to use by protocol or endpoint, e.g., {'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}. The proxies are used on each request.`,name:"proxies"},{anchor:"transformers.GenerationConfig.from_pretrained.token",description:`token (str or bool, optional) — The token to use as HTTP bearer authorization for remote files. If True, or not specified, will use the token generated when running hf auth login (stored in ~/.huggingface).`,name:"token"},{anchor:"transformers.GenerationConfig.from_pretrained.revision",description:`revision (str, optional, defaults to "main") — The specific model version to use. It can be a branch name, a tag name, or a commit id, since we use a git-based system for storing models and other artifacts on huggingface.co, so revision can be any identifier allowed by git.

    To test a pull request you made on the Hub, you can pass revision="refs/pr/<pr_number>".

    `,name:"revision"},{anchor:"transformers.GenerationConfig.from_pretrained.return_unused_kwargs",description:`return_unused_kwargs (bool, optional, defaults to False) — If False, then this function returns just the final configuration object.

    If True, then this functions returns a Tuple(config, unused_kwargs) where unused_kwargs is a dictionary consisting of the key/value pairs whose keys are not configuration attributes: i.e., the part of kwargs which has not been used to update config and is otherwise ignored.`,name:"return_unused_kwargs"},{anchor:"transformers.GenerationConfig.from_pretrained.subfolder",description:`subfolder (str, optional, defaults to "") — In case the relevant files are located inside a subfolder of the model repo on huggingface.co, you can specify the folder name here.`,name:"subfolder"},{anchor:"transformers.GenerationConfig.from_pretrained.kwargs",description:`kwargs (dict[str, Any], optional) — The values in kwargs of any keys which are configuration attributes will be used to override the loaded values. Behavior concerning key/value pairs whose keys are not configuration attributes is controlled by the return_unused_kwargs keyword parameter.`,name:"kwargs"}],source:"https://github.com/huggingface/transformers/blob/v4.56.2/src/transformers/generation/configuration_utils.py#L837",returnDescription:`