wehub-resource-sync
116 lines
4.4 KiB
Python
116 lines
4.4 KiB
Python
import json
|
|
import os
|
|
import multiprocessing
|
|
import itertools
|
|
import random
|
|
|
|
from infinibatch import iterators
|
|
from functools import partial
|
|
from tasks.data.lm_loader import LMLoader
|
|
from tasks.data.utils import NativeCheckpointableIterator, WeightIterator, EOL_SYMBOL, BOI_SYMBOL, EOI_SYMBOL, image_code_to_token
|
|
from fairseq.data.encoders.gpt2_bpe import GPT2BPE
|
|
|
|
|
|
class LaionLoader(LMLoader):
|
|
def _tokenize(self):
|
|
multilingual_iters = []
|
|
weights = []
|
|
|
|
for data in self.data:
|
|
multilingual_iters.append(
|
|
self._tokenize_foreach_lang(data)
|
|
)
|
|
if 'weight' in data:
|
|
weights.append(float(data['weight']))
|
|
else:
|
|
weights.append(int(data['count']))
|
|
|
|
if len(multilingual_iters) == 1:
|
|
return multilingual_iters[0]
|
|
|
|
sampling_iterator = WeightIterator(weights, self.seed)
|
|
control_iterator = NativeCheckpointableIterator(sampling_iterator)
|
|
tokenized_lines = iterators.MultiplexIterator(control_iterator, multilingual_iters)
|
|
|
|
return tokenized_lines
|
|
|
|
def _tokenize_foreach_lang(self, data):
|
|
dataset = list(zip(data['source']))
|
|
if self.shuffle:
|
|
chunk_files = iterators.InfinitePermutationSourceIterator(
|
|
dataset,
|
|
seed=self.seed,
|
|
shuffle=self.shuffle,
|
|
num_instances=self.num_shards,
|
|
instance_rank=self.shard_id,)
|
|
else:
|
|
chunk_files = iterators.ChunkedSourceIterator(
|
|
dataset,
|
|
num_instances=self.num_shards,
|
|
instance_rank=self.shard_id,)
|
|
|
|
tokenized_lines = iterators.SelectManyIterator(chunk_files, lambda files: self._read_from_files(*files))
|
|
tokenized_lines = iterators.SamplingRandomMapIterator(tokenized_lines, self._prepare, self.seed)
|
|
|
|
return tokenized_lines
|
|
|
|
@staticmethod
|
|
def fs_encode_line(fs_dict, words, append_eos=True):
|
|
ids = []
|
|
for i, word in enumerate(words):
|
|
idx = fs_dict.index(word)
|
|
ids.append(idx)
|
|
if append_eos:
|
|
ids.append(fs_dict.eos_index)
|
|
return ids
|
|
|
|
def _read_from_files(self, source_file):
|
|
"""
|
|
<s> <image> image token </image> sentence </s>
|
|
<s> sentence <image> image token </image> </s>
|
|
|
|
"""
|
|
file_path = os.path.join(self.data_dir, source_file)
|
|
|
|
if not os.path.exists(file_path):
|
|
print('| file {} not exists'.format(file_path), flush=True)
|
|
return iter([]) # skip bad file
|
|
|
|
try:
|
|
with open(file_path, 'r', encoding='utf8') as f:
|
|
lines = f.read().strip().split('\n')
|
|
except:
|
|
return iter([]) # skip bad file
|
|
|
|
for doc_jsonstr in lines:
|
|
try:
|
|
obj = json.loads(doc_jsonstr)
|
|
if int(obj['width']) < 200 or int(obj['height']) < 200:
|
|
continue
|
|
|
|
line = obj['caption']
|
|
spm_tokenizer=self.tokenizer
|
|
if isinstance(spm_tokenizer, GPT2BPE):
|
|
tokens = spm_tokenizer.encode(line).split(' ')
|
|
else:
|
|
tokens = spm_tokenizer.encode(line, out_type=str)
|
|
tokenized_tokens = LaionLoader.fs_encode_line(self.dictionary, tokens, append_eos=False)
|
|
image_tokens = [image_code_to_token(i) for i in obj['input_ids']]
|
|
image_tokens = LaionLoader.fs_encode_line(self.dictionary, image_tokens, append_eos=False)
|
|
r = random.random()
|
|
doc = [self.dictionary.bos()]
|
|
if r < 0.5:
|
|
doc.append(self.dictionary.index(BOI_SYMBOL))
|
|
doc.extend(image_tokens)
|
|
doc.append(self.dictionary.index(EOI_SYMBOL))
|
|
doc.extend(tokenized_tokens)
|
|
else:
|
|
doc.extend(tokenized_tokens)
|
|
doc.append(self.dictionary.index(BOI_SYMBOL))
|
|
doc.extend(image_tokens)
|
|
doc.append(self.dictionary.index(EOI_SYMBOL))
|
|
doc.append(self.dictionary.eos())
|
|
yield doc
|
|
except:
|
|
continue
|