requirements.txt

Cython>=0.25
h5py>=2.6,!=3.0.0
numpy>=1.15
pandas>=1.0,!=1.1.5
scipy>=0.18
tabulate>=0.7
scikit-learn
tqdm
torch>=1.13.0
torchaudio
torchtext
torchvision
transformers>=4.28.1
spacy>=2.3
PyYAML>=3.12
absl-py
kaggle
requests
tables
fsspec[http]
dataclasses-json
jsonschema>=4.5.0,<4.7
marshmallow
marshmallow-jsonschema
marshmallow-dataclass==8.5.4
tensorboard
torchmetrics>=0.11
torchinfo
filelock
psutil==5.9.4
protobuf==3.20.3
py-cpuinfo==9.0.0
gpustat
rich~=12.4.4
packaging
retry

# required for TransfoXLTokenizer when using transformer_xl
sacremoses
sentencepiece

# requirements for daft
getdaft

# new data format support
xlwt                # excel
xlrd>=2.0.1         # excel
xlsxwriter>=1.4.3   # excel
openpyxl>=3.0.7     # excel
pyxlsb>=1.0.8       # excel
pyarrow             # parquet
lxml                # html
html5lib            # html