Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Upgrade to tf2.16 #1408

Open
wants to merge 5 commits into
base: main
Choose a base branch
from
Open

Upgrade to tf2.16 #1408

Changes from 1 commit
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
fb23bdf
Upgrade to tf2.16
djherbis Jun 14, 2024
b7793c1
Cleanup tf2.15 => tf2.16
djherbis Jun 14, 2024
b052d13
Remove pins for tf-probability and decision foresets
djherbis Jun 14, 2024
81c132e
tf 2.16.1 actually
djherbis Jun 14, 2024
96abf2d
Bump CUDA to 12.3
djherbis Jun 14, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Prev Previous commit
Next Next commit
Cleanup tf2.15 => tf2.16
  • Loading branch information
@djherbis
djherbis committed Jun 14, 2024
commit b7793c16ad2cfe846d26089c6a2f682a9af97e57
9 changes: 3 additions & 6 deletions Dockerfile.tmpl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -38,9 +38,9 @@ RUN ln -s /usr/local/cuda/lib64/stubs/libcuda.so /usr/local/cuda/lib64/stubs/lib
{{ end }}

# Keep these variables in sync if base image is updated.
ENV TENSORFLOW_VERSION=2.15.0
ENV TENSORFLOW_VERSION=2.16.0
# See https://github.com/tensorflow/io#tensorflow-version-compatibility
ENV TENSORFLOW_IO_VERSION=0.35.0
ENV TENSORFLOW_IO_VERSION=0.37.0

# We need to redefine the ARG here to get the ARG value defined above the FROM instruction.
# See: https://docs.docker.com/engine/reference/builder/#understand-how-arg-and-from-interact
Expand Down Expand Up @@ -220,10 +220,7 @@ RUN chmod +x /tmp/keras_patch.sh && \
ADD patches/keras_internal.py /opt/conda/lib/python3.10/site-packages/tensorflow_decision_forests/keras/keras_internal.py
ADD patches/keras_internal_test.py /opt/conda/lib/python3.10/site-packages/tensorflow_decision_forests/keras/keras_internal_test.py

# Remove "--no-deps" flag and "namex" package once Keras 3.* is included in our base image.
# We ignore dependencies since tf2.15 and Keras 3.* should work despite pip saying it won't.
# Currently, keras tries to install a nightly version of tf 2.16: https://github.com/keras-team/keras/blob/fe2f54aa5bc42fb23a96449cf90434ab9bb6a2cd/requirements.txt#L2
RUN pip install --no-deps "keras>3" keras-cv keras-nlp namex && \
RUN pip install "keras>3" keras-cv keras-nlp && \
/tmp/clean-layer.sh

# b/328788268 libpysal 4.10 seems to fail with "module 'shapely' has no attribute 'Geometry'. Did you mean: 'geometry'"
Expand Down