How to set up TRAX on local computer with win10

Course Q&A Natural Language Processing NLP with Attention Models
1

Is it possible to install trax to work on local computer windows 10?

I am running into errors below after doing “pip install trax” and “pip install resource”

ModuleNotFoundError Traceback (most recent call last)
Input In [1], in <cell line: 6>()
3 import random as rnd
5 # import relevant libraries
----> 6 import trax
7 import trax.fastmath.numpy as np
8 from trax import layers as tl

File ~\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\trax_init_.py:18, in
1 # coding=utf-8
2 # Copyright 2021 The Trax Authors.
3 #
(…)
13 # See the License for the specific language governing permissions and
14 # limitations under the License.
16 “”“Trax top level import.”“”
—> 18 from trax import data
19 from trax import fastmath
20 from trax import layers

File ~\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\trax\data_init_.py:36, in
16 “”“Functions and classes for obtaining and preprocesing data.
17
18 The trax.data module presents a flattened (no subpackages) public API.
(…)
31
32 “””
34 from trax.data.debug_data_pipeline import debug_pipeline
—> 36 from trax.data.inputs import add_loss_weights
37 from trax.data.inputs import addition_inputs
38 from trax.data.inputs import AddLossWeights

File ~\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\trax\data\inputs.py:87, in
84 import numpy as np
85 import tensorflow as tf
—> 87 from trax import fastmath
88 from trax import shapes
89 from trax.data import debug_data_pipeline

File ~\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\trax\fastmath_init_.py:25, in
23 from trax.fastmath.numpy import tree_leaves
24 from trax.fastmath.numpy import tree_unflatten
—> 25 from trax.fastmath.ops import *

File ~\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\trax\fastmath\ops.py:36, in
33 import enum
35 import gin
—> 36 from trax.fastmath.jax import JAX_BACKEND
37 from trax.fastmath.numpy import NUMPY_BACKEND
38 from trax.fastmath.tf import TF_BACKEND

File ~\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\trax\fastmath\jax.py:26, in
24 import numpy as np
25 import tensorflow as tf
—> 26 import tensorflow_datasets as tfds
28 from trax.fastmath import numpy as tnp
29 from trax.shapes import signature

File ~\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\tensorflow_datasets_init_.py:43, in
41 _TIMESTAMP_IMPORT_STARTS = time.time()
42 from absl import logging
—> 43 import tensorflow_datasets.core.logging as _tfds_logging
44 from tensorflow_datasets.core.logging import call_metadata as _call_metadata
46 _metadata = _call_metadata.CallMetadata()

File ~\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\tensorflow_datasets\core_init_.py:22, in
18 # Allow to use tfds.core.Path in dataset implementation which seems more
19 # natural than having to import a third party module.
20 from etils.epath import Path
—> 22 from tensorflow_datasets.core import community
23 from tensorflow_datasets.core.dataset_builder import BeamBasedBuilder
24 from tensorflow_datasets.core.dataset_builder import BuilderConfig

File ~\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\tensorflow_datasets\core\community_init_.py:18, in
1 # coding=utf-8
2 # Copyright 2023 The TensorFlow Datasets Authors.
3 #
(…)
13 # See the License for the specific language governing permissions and
14 # limitations under the License.
16 “”“Community dataset API.”“”
—> 18 from tensorflow_datasets.core.community.huggingface_wrapper import mock_builtin_to_use_gfile
19 from tensorflow_datasets.core.community.huggingface_wrapper import mock_huggingface_import
20 from tensorflow_datasets.core.community.load import builder_cls_from_module

File ~\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\tensorflow_datasets\core\community\huggingface_wrapper.py:31, in
28 from unittest import mock
30 from etils import epath
—> 31 from tensorflow_datasets.core import dataset_builder
32 from tensorflow_datasets.core import dataset_info
33 from tensorflow_datasets.core import download

File ~\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\tensorflow_datasets\core\dataset_builder.py:44, in
42 from tensorflow_datasets.core import reader as reader_lib
43 from tensorflow_datasets.core import registered
—> 44 from tensorflow_datasets.core import split_builder as split_builder_lib
45 from tensorflow_datasets.core import splits as splits_lib
46 from tensorflow_datasets.core import tf_compat

File ~\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\tensorflow_datasets\core\split_builder.py:37, in
35 from tensorflow_datasets.core import splits as splits_lib
36 from tensorflow_datasets.core import utils
—> 37 from tensorflow_datasets.core import writer as writer_lib
38 from tensorflow_datasets.core.utils import shard_utils
40 if typing.TYPE_CHECKING:

File ~\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\tensorflow_datasets\core\writer.py:33, in
31 from tensorflow_datasets.core import lazy_imports_lib
32 from tensorflow_datasets.core import naming
—> 33 from tensorflow_datasets.core import shuffle
34 from tensorflow_datasets.core import utils
35 from tensorflow_datasets.core.utils import file_utils

File ~\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\tensorflow_datasets\core\shuffle.py:20, in
18 import math
19 import os
—> 20 import resource
21 import struct
22 from typing import Iterator, List, Optional

ModuleNotFoundError: No module named ‘resource’

2

Hi @PZ2004

It is possible but it is not easy or straight forward. If you do not have experience with systems administration I would not advise to start from installing trax.

But if you have at least some experience - here is a tutorial(I accidentally linked to wrong tutorial… Sorry :slight_smile: The issue I wanted to link to. )

Cheers

4


1m

Would you comment if it would be easier to use trax in python if I have a linux running on personal computer? would I run into similar configuration problems? Is there a tutorial for linux system?

5

Yes, setting up on linux is way less problematic - tutorial.