Big data? 🤗 Datasets to the rescue!

!pip install datasets evaluate transformers[sentencepiece]

!pip install zstandard

from datasets import load_dataset

# This takes a few minutes to run, so go grab a tea or coffee while you wait :)
data_files = "https://huggingface.co/datasets/qualis2006/PUBMED_title_abstracts_2020_baseline/resolve/main/PUBMED_title_abstracts_2020_baseline.jsonl.zst"
pubmed_dataset = load_dataset("json", data_files=data_files, split="train")
pubmed_dataset

pubmed_dataset[0]

!pip install psutil

import psutil

# Process.memory_info is expressed in bytes, so convert to megabytes
print(f"RAM used: {psutil.Process().memory_info().rss / (1024 * 1024):.2f} MB")

print(f"Number of files in dataset : {pubmed_dataset.dataset_size}")
size_gb = pubmed_dataset.dataset_size / (1024**3)
print(f"Dataset size (cache file) : {size_gb:.2f} GB")

import timeit

code_snippet = """batch_size = 1000

for idx in range(0, len(pubmed_dataset), batch_size):
    _ = pubmed_dataset[idx:idx + batch_size]
"""

time = timeit.timeit(stmt=code_snippet, number=1, globals=globals())
print(
    f"Iterated over {len(pubmed_dataset)} examples (about {size_gb:.1f} GB) in "
    f"{time:.1f}s, i.e. {size_gb/time:.3f} GB/s"
)

pubmed_dataset_streamed = load_dataset(
    "json", data_files=data_files, split="train", streaming=True
)

next(iter(pubmed_dataset_streamed))

from transformers import AutoTokenizer

tokenizer = AutoTokenizer.from_pretrained("distilbert-base-uncased")
tokenized_dataset = pubmed_dataset_streamed.map(lambda x: tokenizer(x["text"]))
next(iter(tokenized_dataset))

shuffled_dataset = pubmed_dataset_streamed.shuffle(buffer_size=10_000, seed=42)
next(iter(shuffled_dataset))

dataset_head = pubmed_dataset_streamed.take(5)
list(dataset_head)

# Skip the first 1,000 examples and include the rest in the training set
train_dataset = shuffled_dataset.skip(1000)
# Take the first 1,000 examples for the validation set
validation_dataset = shuffled_dataset.take(1000)

law_dataset_streamed = load_dataset(
    "json",
    data_files="https://the-eye.eu/public/AI/pile_preliminary_components/FreeLaw_Opinions.jsonl.zst",
    split="train",
    streaming=True,
)
next(iter(law_dataset_streamed))

from itertools import islice
from datasets import interleave_datasets

combined_dataset = interleave_datasets([pubmed_dataset_streamed, law_dataset_streamed])
list(islice(combined_dataset, 2))

base_url = "https://the-eye.eu/public/AI/pile/"
data_files = {
    "train": [base_url + "train/" + f"{idx:02d}.jsonl.zst" for idx in range(30)],
    "validation": base_url + "val.jsonl.zst",
    "test": base_url + "test.jsonl.zst",
}
pile_dataset = load_dataset("json", data_files=data_files, streaming=True)
next(iter(pile_dataset["train"]))