Cherry pick #2241 and #2242 (#2244)

* Remove torchdata dependency from package and from CI (#2241)

* Fix torchdata import error (#2242)

* Remove stuff

* stuff

* lint

---------

Co-authored-by: Nicolas Hug <[email protected]>

Author: huydhn

.circleci/unittest/linux/scripts/install.sh

@@ -1,7 +1,6 @@
 #!/usr/bin/env bash
 
 unset PYTORCH_VERSION
-unset TORCHDATA_VERSION
 # For unittest, nightly PyTorch is used as the following section,
 # so no need to set PYTORCH_VERSION.
 # In fact, keeping PYTORCH_VERSION forces us to hardcode PyTorch version in config.
@@ -30,10 +29,6 @@ printf "* Installing PyTorch\n"
 )
 
 
-printf "Installing torchdata nightly with portalocker\n"
-pip install "portalocker>=2.0.0"
-pip install --pre torchdata --index-url https://download.pytorch.org/whl/nightly/cpu
-
 printf "* Installing torchtext\n"
 python setup.py develop
 

.circleci/unittest/windows/scripts/install.sh

@@ -1,7 +1,6 @@
 #!/usr/bin/env bash
 
 unset PYTORCH_VERSION
-unset TORCHDATA_VERSION
 # For unittest, nightly PyTorch is used as the following section,
 # so no need to set PYTORCH_VERSION.
 # In fact, keeping PYTORCH_VERSION forces us to hardcode PyTorch version in config.
@@ -19,10 +18,6 @@ conda activate ./env
 printf "* Installing PyTorch\n"
 conda install -y -c "pytorch-${UPLOAD_CHANNEL}" ${CONDA_CHANNEL_FLAGS} pytorch cpuonly
 
-printf "* Installing torchdata nightly with portalocker\n"
-pip install "portalocker>=2.0.0"
-pip install --pre torchdata --index-url https://download.pytorch.org/whl/nightly/cpu
-
 printf "* Installing pywin32_postinstall script\n"
 curl --output pywin32_postinstall.py https://raw.githubusercontent.com/mhammond/pywin32/main/pywin32_postinstall.py
 python pywin32_postinstall.py -install

.github/workflows/build-conda-linux.yml

@@ -29,7 +29,7 @@ jobs:
       matrix:
         include:
           - repository: pytorch/text
-            pre-script: packaging/install_torchdata.sh
+            pre-script: ""
             post-script: ""
             conda-package-directory: packaging/torchtext
             smoke-test-script: test/smoke_tests/smoke_tests.py

.github/workflows/build-conda-m1.yml

@@ -28,7 +28,7 @@ jobs:
       matrix:
         include:
           - repository: pytorch/text
-            pre-script: packaging/install_torchdata.sh
+            pre-script: ""
             post-script: ""
             conda-package-directory: packaging/torchtext
             smoke-test-script: test/smoke_tests/smoke_tests.py

.github/workflows/build-conda-windows.yml

@@ -29,7 +29,7 @@ jobs:
       matrix:
         include:
           - repository: pytorch/text
-            pre-script: packaging/install_torchdata.sh
+            pre-script: ""
             post-script: ""
             conda-package-directory: packaging/torchtext
             smoke-test-script: test/smoke_tests/smoke_tests.py

.github/workflows/build-wheels-linux.yml

@@ -34,7 +34,7 @@ jobs:
       matrix:
         include:
           - repository: pytorch/text
-            pre-script: packaging/install_torchdata.sh
+            pre-script: ""
             post-script: ""
             smoke-test-script: test/smoke_tests/smoke_tests.py
             package-name: torchtext

.github/workflows/build-wheels-m1.yml

@@ -32,7 +32,7 @@ jobs:
       matrix:
         include:
           - repository: pytorch/text
-            pre-script: packaging/install_torchdata.sh
+            pre-script: ""
             post-script: ""
             package-name: torchtext
             smoke-test-script: test/smoke_tests/smoke_tests.py

.github/workflows/build-wheels-windows.yml

@@ -33,7 +33,7 @@ jobs:
       matrix:
         include:
           - repository: pytorch/text
-            pre-script: packaging/install_torchdata.sh
+            pre-script: ""
             env-script: packaging/vc_env_helper.bat
             post-script: ""
             smoke-test-script: test/smoke_tests/smoke_tests.py

.github/workflows/codeql.yml

@@ -31,7 +31,6 @@ jobs:
       - name: Install Torch
         run: |
           python -m pip install cmake
-          python -m pip install --quiet --pre torch torchdata -f https://download.pytorch.org/whl/nightly/cpu/torch_nightly.html
           sudo ln -s /usr/bin/ninja /usr/bin/ninja-build
 
       - name: Build TorchText

.github/workflows/integration-test.yml

@@ -39,15 +39,13 @@ jobs:
         python -m spacy download en_core_web_sm
         printf "* Downloading SpaCy German models\n"
         python -m spacy download de_core_news_sm
-        # Install PyTorch, Torchvision, and TorchData
+        # Install PyTorch, Torchvision
         set -ex
         conda install \
           --yes \
           -c "pytorch-${CHANNEL}" \
           -c nvidia "pytorch-${CHANNEL}"::pytorch[build="*${VERSION}*"] \
           "${CUDATOOLKIT}"
-        printf "Installing torchdata nightly\n"
-        python3 -m pip install --pre torchdata --index-url https://download.pytorch.org/whl/nightly/cpu
         python3 setup.py develop
         # Install integration test dependencies
         python3 -m pip --quiet install parameterized

.github/workflows/test-linux-cpu.yml

@@ -50,16 +50,13 @@ jobs:
         printf "* Downloading SpaCy German models\n"
         python -m spacy download de_core_news_sm
 
-        # Install PyTorch, Torchvision, and TorchData
+        # Install PyTorch, Torchvision
         set -ex
         conda install \
           --yes \
           -c "pytorch-${CHANNEL}" \
           -c nvidia "pytorch-${CHANNEL}"::pytorch[build="*${VERSION}*"] \
           "${CUDATOOLKIT}"
-        printf "Installing torchdata nightly\n"
-        python3 -m pip install "portalocker>=2.0.0"
-        python3 -m pip install --pre torchdata --index-url https://download.pytorch.org/whl/nightly/cpu
         python3 setup.py develop
         python3 -m pip install parameterized
 

.github/workflows/test-linux-gpu.yml

@@ -54,17 +54,14 @@ jobs:
         printf "* Downloading SpaCy German models\n"
         python -m spacy download de_core_news_sm
 
-        # Install PyTorch and TorchData
+        # Install PyTorch
         set -ex
         conda install \
           --yes \
           --quiet \
           -c "pytorch-${CHANNEL}" \
           -c nvidia "pytorch-${CHANNEL}"::pytorch[build="*${VERSION}*"] \
           "${CUDATOOLKIT}"
-        printf "Installing torchdata nightly\n"
-        python3 -m pip install "portalocker>=2.0.0"
-        python3 -m pip install --pre torchdata --index-url https://download.pytorch.org/whl/nightly/cpu --quiet
         python3 setup.py develop
         python3 -m pip install parameterized --quiet
 

.github/workflows/test-macos-cpu.yml

@@ -55,7 +55,7 @@ jobs:
         printf "* Downloading SpaCy German models\n"
         python -m spacy download de_core_news_sm
 
-        # Install PyTorch, Torchvision, and TorchData
+        # Install PyTorch, Torchvision
         set -ex
         conda install \
           --yes \
@@ -64,9 +64,6 @@ jobs:
           "${MKL_CONSTRAINT}" \
           pytorch \
           "${CUDATOOLKIT}"
-        printf "Installing torchdata nightly\n"
-        python3 -m pip install "portalocker>=2.0.0"
-        python3 -m pip install --pre torchdata --index-url https://download.pytorch.org/whl/nightly/cpu
         python3 setup.py develop
         python3 -m pip install parameterized
 

.github/workflows/test-windows-cpu.yml

@@ -51,15 +51,12 @@ jobs:
         printf "* Downloading SpaCy German models\n"
         python -m spacy download de_core_news_sm
 
-        # Install PyTorch, Torchvision, and TorchData
+        # Install PyTorch, Torchvision
         conda install \
           --yes \
           -c "pytorch-${CHANNEL}" \
           pytorch \
           cpuonly
-        printf "Installing torchdata nightly\n"
-        python -m pip install "portalocker>=2.0.0"
-        python -m pip install --pre torchdata --index-url https://download.pytorch.org/whl/nightly/cpu
 
         printf "* Installing pywin32_postinstall script\n"
         curl --output pywin32_postinstall.py https://raw.githubusercontent.com/mhammond/pywin32/main/pywin32_postinstall.py

.github/workflows/validate-binaries.yml

@@ -43,6 +43,11 @@ on:
         default: ""
         required: false
         type: string
+      pytorch_version:
+        description: "PyTorch version to validate (ie. 2.0, 2.2.2, etc.) - optional"
+        default: ""
+        required: false
+        type: string
 jobs:
   validate-binaries:
     uses: pytorch/test-infra/.github/workflows/validate-domain-library.yml@release/2.2

README.rst

@@ -12,6 +12,9 @@
 torchtext
 +++++++++
 
+CAUTION: As of September 2023 we have paused active development of TorchText because our focus has shifted away from building out this library offering.
+We will continue to release new versions but do not anticipate any new feature development as we figure out future investments in this space.
+
 This repository consists of:
 
 * `torchtext.datasets <https://github.com/pytorch/text/tree/main/torchtext/datasets>`_: The raw text iterators for common NLP datasets

packaging/install_torchdata.sh

@@ -190,14 +190,6 @@ setup_pip_pytorch_version() {
       -f https://download.pytorch.org/whl/torch_stable.html \
       -f "https://download.pytorch.org/whl/${UPLOAD_CHANNEL}/torch_${UPLOAD_CHANNEL}.html"
   fi
-  if [[ -z "$TORCHDATA_VERSION" ]]; then
-    pip_install --pre torchdata -f "https://download.pytorch.org/whl/nightly/cpu/torch_nightly.html"
-    export TORCHDATA_VERSION="$(pip show torchdata | grep ^Version: | sed 's/Version:  *//' | sed 's/+.\+//')"
-  else
-    pip_install "torchdata==$TORCHDATA_VERSION" \
-      -f https://download.pytorch.org/whl/torch_stable.html \
-      -f "https://download.pytorch.org/whl/${UPLOAD_CHANNEL}/torch_${UPLOAD_CHANNEL}.html"
-  fi
 }
 
 # Fill PYTORCH_VERSION with the latest conda nightly version, and
@@ -232,10 +224,6 @@ setup_conda_pytorch_constraint() {
       export CONDA_EXTRA_BUILD_CONSTRAINT="- mkl<=2021.2.0"
     fi
   fi
-  if [[ -z "$TORCHDATA_VERSION" ]]; then
-    export TORCHDATA_VERSION="$(conda search --json 'torchdata[channel=pytorch-nightly]' | ${PYTHON} -c "import sys, json, re; print(re.sub(r'\\+.*$', '', json.load(sys.stdin)['torchdata'][-1]['version']))")"
-  fi
-  export CONDA_TORCHDATA_CONSTRAINT="- torchdata==$TORCHDATA_VERSION"
 }
 
 # Translate CUDA_VERSION into CUDA_CUDATOOLKIT_CONSTRAINT

packaging/pkg_helpers.bash

@@ -24,7 +24,6 @@ requirements:
     - requests
     - tqdm
     {{ environ.get('CONDA_PYTORCH_CONSTRAINT') }}
-    {{ environ.get('CONDA_TORCHDATA_CONSTRAINT') }}
 
 build:
   string: py{{py}}

packaging/torchtext/meta.yaml

@@ -1,4 +1,5 @@
 [pytest]
+addopts = --ignore-glob=test/torchtext_unittest/datasets/*
 testpaths = test/
 python_paths = ./
 markers =

pytest.ini

@@ -19,7 +19,6 @@ Sphinx
 pytest
 expecttest
 parameterized
-torchdata>0.5
 
 # Lets pytest find our code by automatically modifying PYTHONPATH
 pytest-pythonpath

requirements.txt

@@ -63,14 +63,10 @@ def _init_submodule():
 print("-- Building version " + VERSION)
 
 pytorch_package_version = os.getenv("PYTORCH_VERSION")
-torchdata_package_version = os.getenv("TORCHDATA_VERSION")
 
 pytorch_package_dep = "torch"
 if pytorch_package_version is not None:
     pytorch_package_dep += "==" + pytorch_package_version
-torchdata_package_dep = "torchdata"
-if torchdata_package_version is not None:
-    torchdata_package_dep += "==" + torchdata_package_version
 
 
 class clean(distutils.command.clean.clean):
@@ -104,7 +100,7 @@ def run(self):
     description="Text utilities, models, transforms, and datasets for PyTorch.",
     long_description=read("README.rst"),
     license="BSD",
-    install_requires=["tqdm", "requests", pytorch_package_dep, "numpy", torchdata_package_dep],
+    install_requires=["tqdm", "requests", pytorch_package_dep, "numpy"],
     python_requires=">=3.8",
     classifiers=[
         "Programming Language :: Python :: 3.8",

setup.py

@@ -1,28 +1,6 @@
 """Run smoke tests"""
 
-import os
-import re
-
-import torchdata
 import torchtext
-import torchtext.version  # noqa: F401
-
-NIGHTLY_ALLOWED_DELTA = 3
-channel = os.getenv("MATRIX_CHANNEL")
-
-
-def validateTorchdataVersion():
-    from datetime import datetime
-
-    date_t_str = re.findall(r"dev\d+", torchdata.__version__)[0]
-    date_t_delta = datetime.now() - datetime.strptime(date_t_str[3:], "%Y%m%d")
-
-    if date_t_delta.days >= NIGHTLY_ALLOWED_DELTA:
-        raise RuntimeError(f"torchdata binary {torchdata.__version__} is more than {NIGHTLY_ALLOWED_DELTA} days old!")
-
 
-if channel == "nightly":
-    validateTorchdataVersion()
 
 print("torchtext version is ", torchtext.__version__)
-print("torchdata version is ", torchdata.__version__)

test/smoke_tests/smoke_tests.py

@@ -4,7 +4,6 @@
 
 # This is to allow monkey-patching in fbcode
 from torch.hub import load_state_dict_from_url  # noqa
-from torchdata.datapipes.iter import HttpReader, GDriveReader  # noqa F401
 from tqdm import tqdm
 
 

torchtext/_download_hooks.py

@@ -2,8 +2,6 @@
 from functools import partial
 from typing import Union, Tuple
 
-from torchdata.datapipes.iter import FileOpener, IterableWrapper
-from torchtext._download_hooks import HttpReader
 from torchtext._internal.module_utils import is_module_available
 from torchtext.data.datasets_utils import (
     _wrap_split_argument,
@@ -65,6 +63,7 @@ def AG_NEWS(root: str, split: Union[Tuple[str], str]):
         raise ModuleNotFoundError(
             "Package `torchdata` not found. Please install following instructions at https://github.com/pytorch/data"
         )
+    from torchdata.datapipes.iter import FileOpener, GDriveReader, HttpReader, IterableWrapper  # noqa
 
     url_dp = IterableWrapper([URL[split]])
     cache_dp = url_dp.on_disk_cache(

torchtext/datasets/ag_news.py

@@ -2,8 +2,6 @@
 from functools import partial
 from typing import Union, Tuple
 
-from torchdata.datapipes.iter import FileOpener, IterableWrapper
-from torchtext._download_hooks import GDriveReader
 from torchtext._internal.module_utils import is_module_available
 from torchtext.data.datasets_utils import (
     _wrap_split_argument,
@@ -79,6 +77,7 @@ def AmazonReviewFull(root: str, split: Union[Tuple[str], str]):
         raise ModuleNotFoundError(
             "Package `torchdata` not found. Please install following instructions at https://github.com/pytorch/data"
         )
+    from torchdata.datapipes.iter import FileOpener, GDriveReader, HttpReader, IterableWrapper  # noqa
 
     url_dp = IterableWrapper([URL])
     cache_compressed_dp = url_dp.on_disk_cache(