def test_tokenizer(test_sentence, vocab_path, merge_path): r""" Illustrates how the individual Tokenizer works Args: test_sentence (:obj:`str`): Sentence for demonstration purposes vocab_path (:obj:`str`): Path where the vocabulary (most frequent tokens ranked by frequency) is saved merge_path (:obj:`str`): Path where the merges file is saved """ tokenizer = ByteLevelBPETokenizer(vocab_path, merge_path) tokenizer._tokenizer.post_processor = BertProcessing( ("</s>", tokenizer.token_to_id("</s>")), ("<s>", tokenizer.token_to_id("<s>"))) tokenizer.enable_truncation(max_length=512) print("Original sentence " + test_sentence) print("Encoded string: {}".format(tokenizer.encode(test_sentence).tokens)) encoding = tokenizer.encode(test_sentence) decoded = tokenizer.decode(encoding.ids) print("Decoded string: {}".format(decoded))
def train_tokenizer(data_path, wiki_text_file_path): # ToDo := Load if weights exists, else setup tokenizer_en = GPT2TokenizerFast.from_pretrained("gpt2") tokenizer_en.pad_token = tokenizer_en.eos_token vocab_size = tokenizer_en.vocab_size max_length = 1024 tokenizer_es = ByteLevelBPETokenizer() tokenizer_es.train( files=[str(wiki_text_file_path)], vocab_size=vocab_size, min_frequency=2, special_tokens=[EOF_TOKEN] ) tokenizer_es.enable_truncation(max_length=max_length) tokenizer_es_path = data_path/"BLBPE_tokenizer_es" tokenizer_es_path.mkdir(exist_ok=True, parents=True) tokenizer_es.save_model(str(tokenizer_es_path)) tokenizer_es = GPT2TokenizerFast.from_pretrained( str(tokenizer_es_path), pad_token=EOF_TOKEN ) tokenizer_es.model_max_length = max_length # tokenizer_es = ByteLevelBPETokenizer( # vocab_file=str(tokenizer_es_path/"vocab.json"), # merges_file=str(tokenizer_es_path/"merges.txt"), # ) # tokenizer_es.enable_truncation(max_length=1024) # ToDo := is this necessary # tokenizer_en.pad_token = tokenizer_en.eos_token return tokenizer_en, tokenizer_es
class LineByLineTextDataset(Dataset): def __init__(self, args, file_path: str, block_size=512): assert os.path.isfile(file_path) self.block_size = block_size self.tokenizer = ByteLevelBPETokenizer( os.path.join(args.tokenizer_name, "vocab.json"), os.path.join(args.tokenizer_name, "merges.txt"), ) self.tokenizer._tokenizer.post_processor = RobertaProcessing( ("</s>", self.tokenizer.token_to_id("</s>")), ("<s>", self.tokenizer.token_to_id("<s>")), ) self.tokenizer.enable_truncation(max_length=block_size) logger.info("Creating features from dataset file at %s", file_path) self.examples = [] with open(file_path, encoding="utf-8") as f: for line in f: if len(line) > 0 and not line.isspace(): self.examples.append(line) def __len__(self): return len(self.examples) def __getitem__(self, i): return torch.tensor(self.tokenizer.encode(self.examples[i]).ids[: self.block_size - 2], dtype=torch.long)
def __init__(self, evaluate: bool = False): tokenizer = ByteLevelBPETokenizer( "./model/bbpe/vocab.json", "./model/bbpe/merges.txt", ) tokenizer._tokenizer.post_processor = BertProcessing( ("</s>", tokenizer.token_to_id("</s>")), ("<s>", tokenizer.token_to_id("<s>")), ) tokenizer.enable_truncation(max_length=512) # or use the RobertaTokenizer from `transformers` directly. self.examples = [] src_files = Path("./data/").glob("*_eval.csv") if evaluate else Path( "./data/").glob("*_eval.csv") for src_file in src_files: print("🔥", src_file) with open(src_file, 'r', encoding='utf-8') as f: for index, line in enumerate(f): self.examples += [ x.ids for x in tokenizer.encode_batch(line) ] if index % 10000 == 0: print(src_file, index // 10000)
def load_sentence_piece_model(): tokenizer = ByteLevelBPETokenizer(path_vocab, path_model) tokenizer._tokenizer.post_processor = BertProcessing( ("</s>", tokenizer.token_to_id("</s>")), ("<s>", tokenizer.token_to_id("<s>"))) tokenizer.enable_truncation(max_length=512) encoding = tokenizer.encode("배고파요") print(encoding.tokens) print(encoding.special_tokens_mask) print(encoding.ids) print(encoding.normalized_str)
def __init__(self, max_tokens=512): ## RoBERTa uses BPE tokenizer similar to GPT t = ByteLevelBPETokenizer("tokenizer/vocab.json", "tokenizer/merges.txt") t._tokenizer.post_processor = BertProcessing( ("</s>", t.token_to_id("</s>")), ("<s>", t.token_to_id("<s>")), ) t.enable_truncation(max_tokens) t.enable_padding(length=max_tokens, pad_id=t.token_to_id("<pad>")) self.tokenizer = t
def load_sentence_piece_model(path_vocab, path_model): tokenizer = ByteLevelBPETokenizer(path_vocab, path_model) tokenizer._tokenizer.post_processor = BertProcessing( ("<bos>", tokenizer.token_to_id("<bos>")), ("<eos>", tokenizer.token_to_id("<eos>")) ) tokenizer.enable_truncation(max_length=512) # encoding = tokenizer.encode("배고파요") # print(encoding.tokens) # print(encoding.special_tokens_mask) # print(encoding.ids) # print(encoding.normalized_str) # # decoding = tokenizer.decode([2, 1177, 276, 692, 571, 1]) # print(decoding) return tokenizer
def __init__(self, evaluate: bool = False): tokenizer = ByteLevelBPETokenizer( "models/faberto/vocab.json", "models/faberto/merges.txt", ) tokenizer._tokenizer.post_processor = BertProcessing( ("</s>", tokenizer.token_to_id("</s>")), ("<s>", tokenizer.token_to_id("<s>")), ) tokenizer.enable_truncation(max_length=512) # or use the RobertaTokenizer from `transformers` directly. self.examples = [] src_files = Path("./data/").glob("*-eval.txt") if evaluate else Path( "./data/").glob("*-train.txt") for src_file in src_files: print("🔥", src_file) lines = src_file.open(encoding="utf-8").read().splitlines() self.examples += [x.ids for x in tokenizer.encode_batch(lines)]
def inference(): from tokenizers import ByteLevelBPETokenizer from tokenizers.processors import BertProcessing ''' initialize tokenizer with saved model files ''' tokenizer = ByteLevelBPETokenizer( "./tok_checkpoints/tokenizer_model-vocab.json", "./tok_checkpoints/tokenizer_model-merges.txt", ) ''' optional step : preprocess the strings Ex: add <s> and </s> as BOS and EOS tokens to the string pad string to some max length and truncate string to some max length ''' tokenizer._tokenizer.post_processor = BertProcessing( ("</s>", tokenizer.token_to_id("</s>")), ("<s>", tokenizer.token_to_id("<s>")), ) tokenizer.enable_padding(pad_token='<pad>', pad_id=tokenizer.get_vocab()['<pad>'], length=20) tokenizer.enable_truncation(max_length=20) ''' tokenize/encode strings ''' input_ids = tokenizer.encode("Hello World, Whats up!!!").ids print("input ids", input_ids) tokens = tokenizer.encode("Hello World, Whats up!!!").tokens print("tokens", tokens) ''' tokenize/encode batch of string ''' batch_tokenized = tokenizer.encode_batch( ["Hello World, Whats up!!!", "Whata whata wa wada wada"]) input_ids = [i.ids for i in batch_tokenized] print("input ids", input_ids) tokens = [i.tokens for i in batch_tokenized] print("tokens", tokens)
def __init__(self, evaluate: bool = False): tokenizer = ByteLevelBPETokenizer( vocab_file, merges_file ) tokenizer._tokenizer.post_processor = BertProcessing( ("</s>", tokenizer.token_to_id("</s>")), ("<s>", tokenizer.token_to_id("<s>")), ) tokenizer.enable_truncation(max_length=512) self.examples = [] src_files = Path(data_folder).glob("**/*.txt") for src_file in src_files: print("🇩🇰", src_file) lines = src_file.read_text(encoding="utf-8").splitlines() self.examples += [x.ids for x in tokenizer.encode_batch(lines)]
class ByteBPETokenizer: def __init__(self, vocab_json, merge_txt, max_length=750): self.tokenizer = ByteLevelBPETokenizer(vocab_json, merge_txt) self.tokenizer.enable_truncation(max_length=max_length) self.tokenizer.enable_padding(max_length=max_length) self.tokenizer.add_special_tokens(["[PAD]", "[CLS]"]) # self.tokenizer.post_processor = RobertaProcessing(("</s>", 2), ("<s>", 1)) # self.tokenizer = RobertaTokenizer.from_pretrained('roberta-base') def encode(self, review): review = clean_sentence(review) encoded = self.tokenizer.encode(review.lower()) # pp_encoded = self.tokenizer.post_process(encoded) return encoded def tokenize2Index(self, review, should_stem=False): encoded = self.encode(review) return encoded.ids def trainBPE(self, paths, vocab_size=30000, min_frequency=10, special_tokens=["[PAD]", "[CLS]"]): tokenizer = ByteLevelBPETokenizer() tokenizer.train(files=paths, vocab_size=vocab_size, min_frequency=min_frequency, special_tokens=special_tokens) tokenizer.save("yelp_bpe/", "yelp-bpe")
# Customize training tokenizer.train(files=paths, vocab_size=52_000, min_frequency=2, special_tokens=[ "<s>", "<pad>", "</s>", "<unk>", "<mask>", ]) # Save files to disk tokenizer.save(".", "rubinberto") tokenizer = ByteLevelBPETokenizer( "rubinberto-vocab.json", "rubinberto-merges.txt", ) tokenizer._tokenizer.post_processor = BertProcessing( ("</s>", tokenizer.token_to_id("</s>")), ("<s>", tokenizer.token_to_id("<s>")), ) tokenizer.enable_truncation(max_length=512) print( tokenizer.encode( "А можно вспоминать не о событиях, а, например, о чувствах, испытываемых нами за «отчетный период»." ).tokens)
def main(): parser = argparse.ArgumentParser() # Required parameters parser.add_argument( "--train_data_file", default=None, type=str, required=True, help="The input training data file (a text file)." ) parser.add_argument( "--output_dir", type=str, required=True, help="The output directory where the model predictions and checkpoints will be written.", ) parser.add_argument( "--model_type", type=str, required=True, help="The model architecture to be trained or fine-tuned.", ) # Other parameters parser.add_argument( "--eval_data_file", default=None, type=str, help="An optional input evaluation data file to evaluate the perplexity on (a text file).", ) parser.add_argument( "--line_by_line", action="store_true", help="Whether distinct lines of text in the dataset are to be handled as distinct sequences.", ) parser.add_argument( "--should_continue", action="store_true", help="Whether to continue from latest checkpoint in output_dir" ) parser.add_argument( "--model_name_or_path", default=None, type=str, help="The model checkpoint for weights initialization. Leave None if you want to train a model from scratch.", ) parser.add_argument( "--mlm", action="store_true", help="Train with masked-language modeling loss instead of language modeling." ) parser.add_argument( "--mlm_probability", type=float, default=0.15, help="Ratio of tokens to mask for masked language modeling loss" ) parser.add_argument( "--config_name", default=None, type=str, help="Optional pretrained config name or path if not the same as model_name_or_path. If both are None, initialize a new config.", ) parser.add_argument( "--tokenizer_name", default=None, type=str, help="Optional pretrained tokenizer name or path if not the same as model_name_or_path. If both are None, initialize a new tokenizer.", ) parser.add_argument( "--cache_dir", default=None, type=str, help="Optional directory to store the pre-trained models downloaded from s3 (instead of the default one)", ) parser.add_argument( "--block_size", default=-1, type=int, help="Optional input sequence length after tokenization." "The training dataset will be truncated in block of this size for training." "Default to the model max input length for single sentence inputs (take into account special tokens).", ) parser.add_argument("--do_train", action="store_true", help="Whether to run training.") parser.add_argument("--do_eval", action="store_true", help="Whether to run eval on the dev set.") parser.add_argument( "--evaluate_during_training", action="store_true", help="Run evaluation during training at each logging step." ) parser.add_argument("--per_gpu_train_batch_size", default=4, type=int, help="Batch size per GPU/CPU for training.") parser.add_argument( "--per_gpu_eval_batch_size", default=4, type=int, help="Batch size per GPU/CPU for evaluation." ) parser.add_argument( "--gradient_accumulation_steps", type=int, default=1, help="Number of updates steps to accumulate before performing a backward/update pass.", ) parser.add_argument("--learning_rate", default=5e-5, type=float, help="The initial learning rate for Adam.") parser.add_argument("--weight_decay", default=0.0, type=float, help="Weight decay if we apply some.") parser.add_argument("--adam_epsilon", default=1e-8, type=float, help="Epsilon for Adam optimizer.") parser.add_argument("--max_grad_norm", default=1.0, type=float, help="Max gradient norm.") parser.add_argument( "--num_train_epochs", default=1.0, type=float, help="Total number of training epochs to perform." ) parser.add_argument( "--max_steps", default=-1, type=int, help="If > 0: set total number of training steps to perform. Override num_train_epochs.", ) # parser.add_argument("--warmup_steps", default=0, type=int, help="Linear warmup over warmup_steps.") parser.add_argument("--pct_warmup", default=0.3, type=float, help="Linear warmup over pct_warmup * total_steps.") parser.add_argument("--logging_steps", type=int, default=500, help="Log every X updates steps.") parser.add_argument("--save_steps", type=int, default=500, help="Save checkpoint every X updates steps.") parser.add_argument( "--save_total_limit", type=int, default=None, help="Limit the total amount of checkpoints, delete the older checkpoints in the output_dir, does not delete by default", ) parser.add_argument( "--eval_all_checkpoints", action="store_true", help="Evaluate all checkpoints starting with the same prefix as model_name_or_path ending and ending with step number", ) parser.add_argument("--no_cuda", action="store_true", help="Avoid using CUDA when available") parser.add_argument( "--overwrite_output_dir", action="store_true", help="Overwrite the content of the output directory" ) parser.add_argument( "--overwrite_cache", action="store_true", help="Overwrite the cached training and evaluation sets" ) parser.add_argument("--seed", type=int, default=42, help="random seed for initialization") parser.add_argument( "--fp16", action="store_true", help="Whether to use 16-bit (mixed) precision (through NVIDIA apex) instead of 32-bit", ) parser.add_argument( "--fp16_opt_level", type=str, default="O1", help="For fp16: Apex AMP optimization level selected in ['O0', 'O1', 'O2', and 'O3']." "See details at https://nvidia.github.io/apex/amp.html", ) parser.add_argument("--local_rank", type=int, default=-1, help="For distributed training: local_rank") parser.add_argument("--server_ip", type=str, default="", help="For distant debugging.") parser.add_argument("--server_port", type=str, default="", help="For distant debugging.") args = parser.parse_args() if args.model_type in ["bert", "roberta", "distilbert", "camembert"] and not args.mlm: raise ValueError( "BERT and RoBERTa-like models do not have LM heads but masked LM heads. They must be run using the --mlm " "flag (masked language modeling)." ) if args.eval_data_file is None and args.do_eval: raise ValueError( "Cannot do evaluation without an evaluation data file. Either supply a file to --eval_data_file " "or remove the --do_eval argument." ) if args.should_continue: sorted_checkpoints = _sorted_checkpoints(args) if len(sorted_checkpoints) == 0: raise ValueError("Used --should_continue but no checkpoint was found in --output_dir.") else: args.model_name_or_path = sorted_checkpoints[-1] if ( os.path.exists(args.output_dir) and os.listdir(args.output_dir) and args.do_train and not args.overwrite_output_dir ): raise ValueError( "Output directory ({}) already exists and is not empty. Use --overwrite_output_dir to overcome.".format( args.output_dir ) ) # Setup distant debugging if needed if args.server_ip and args.server_port: # Distant debugging - see https://code.visualstudio.com/docs/python/debugging#_attach-to-a-local-script import ptvsd print("Waiting for debugger attach") ptvsd.enable_attach(address=(args.server_ip, args.server_port), redirect_output=True) ptvsd.wait_for_attach() # Setup CUDA, GPU & distributed training if args.local_rank == -1 or args.no_cuda: device = torch.device("cuda" if torch.cuda.is_available() and not args.no_cuda else "cpu") args.n_gpu = 0 if args.no_cuda else torch.cuda.device_count() else: # Initializes the distributed backend which will take care of sychronizing nodes/GPUs torch.cuda.set_device(args.local_rank) device = torch.device("cuda", args.local_rank) torch.distributed.init_process_group(backend="nccl") args.n_gpu = 1 args.device = device # Setup logging logging.basicConfig( format="%(asctime)s - %(levelname)s - %(name)s - %(message)s", datefmt="%m/%d/%Y %H:%M:%S", level=logging.INFO if args.local_rank in [-1, 0] else logging.WARN, ) logger.warning( "Process rank: %s, device: %s, n_gpu: %s, distributed training: %s, 16-bits training: %s", args.local_rank, device, args.n_gpu, bool(args.local_rank != -1), args.fp16, ) # Set seed set_seed(args) # Load pretrained model and tokenizer if args.local_rank not in [-1, 0]: torch.distributed.barrier() # Barrier to make sure only the first process in distributed training download model & vocab config_class, model_class, tokenizer_class = MODEL_CLASSES[args.model_type] if args.config_name: config = config_class.from_pretrained(args.config_name, cache_dir=args.cache_dir) elif args.model_name_or_path: config = config_class.from_pretrained(args.model_name_or_path, cache_dir=args.cache_dir) else: config = config_class() if args.tokenizer_name: # tokenizer = tokenizer_class.from_pretrained(args.tokenizer_name, cache_dir=args.cache_dir) tokenizer = ByteLevelBPETokenizer( os.path.join(args.tokenizer_name, "vocab.json"), os.path.join(args.tokenizer_name, "merges.txt") ) tokenizer._tokenizer.post_processor = BertProcessing( ("</s>", tokenizer.token_to_id("</s>")), ("<s>", tokenizer.token_to_id("<s>")), ) tokenizer.enable_truncation(max_length=512) elif args.model_name_or_path: tokenizer = tokenizer_class.from_pretrained(args.model_name_or_path, cache_dir=args.cache_dir) else: raise ValueError( "You are instantiating a new {} tokenizer. This is not supported, but you can do it from another script, save it," "and load it from here, using --tokenizer_name".format(tokenizer_class.__name__) ) # if args.block_size <= 0: # args.block_size = tokenizer.max_len # # Our input block size will be the max possible for the model # else: # args.block_size = min(args.block_size, tokenizer.max_len) if args.model_name_or_path: model = model_class.from_pretrained( args.model_name_or_path, from_tf=bool(".ckpt" in args.model_name_or_path), config=config, cache_dir=args.cache_dir, ) else: logger.info("Training new model from scratch") model = model_class(config=config) model.to(args.device) if args.local_rank == 0: torch.distributed.barrier() # End of barrier to make sure only the first process in distributed training download model & vocab logger.info("Training/evaluation parameters %s", args) # Training if args.do_train: if args.local_rank not in [-1, 0]: torch.distributed.barrier() # Barrier to make sure only the first process in distributed training process the dataset, and the others will use the cache train_dataset = load_and_cache_examples(args, tokenizer, evaluate=False) if args.local_rank == 0: torch.distributed.barrier() global_step, tr_loss = train(args, train_dataset, model, tokenizer) logger.info(" global_step = %s, average loss = %s", global_step, tr_loss) # Saving best-practices: if you use save_pretrained for the model and tokenizer, you can reload them using from_pretrained() if args.do_train and (args.local_rank == -1 or torch.distributed.get_rank() == 0): # Create output directory if needed if args.local_rank in [-1, 0]: os.makedirs(args.output_dir, exist_ok=True) logger.info("Saving model checkpoint to %s", args.output_dir) # Save a trained model, configuration and tokenizer using `save_pretrained()`. # They can then be reloaded using `from_pretrained()` model_to_save = ( model.module if hasattr(model, "module") else model ) # Take care of distributed/parallel training model_to_save.save_pretrained(args.output_dir) tokenizer.save_pretrained(args.output_dir) # Good practice: save your training arguments together with the trained model torch.save(args, os.path.join(args.output_dir, "training_args.bin")) # Load a trained model and vocabulary that you have fine-tuned model = model_class.from_pretrained(args.output_dir) tokenizer = tokenizer_class.from_pretrained(args.output_dir) model.to(args.device) # Evaluation results = {} if args.do_eval and args.local_rank in [-1, 0]: checkpoints = [args.output_dir] if args.eval_all_checkpoints: checkpoints = list( os.path.dirname(c) for c in sorted(glob.glob(args.output_dir + "/**/" + WEIGHTS_NAME, recursive=True)) ) logging.getLogger("transformers.modeling_utils").setLevel(logging.WARN) # Reduce logging logger.info("Evaluate the following checkpoints: %s", checkpoints) for checkpoint in checkpoints: global_step = checkpoint.split("-")[-1] if len(checkpoints) > 1 else "" prefix = checkpoint.split("/")[-1] if checkpoint.find("checkpoint") != -1 else "" model = model_class.from_pretrained(checkpoint) model.to(args.device) result = evaluate(args, model, tokenizer, prefix=prefix) result = dict((k + "_{}".format(global_step), v) for k, v in result.items()) results.update(result) return results