chriscummins / phd Goto Github PK

👨‍💻 My PhD.

C++ 30.21% C 27.99% Python 19.94% Shell 0.90% TeX 0.12% Jupyter Notebook 14.70% JavaScript 0.03% Makefile 3.28% Perl 0.14% CoffeeScript 0.01% CSS 0.01% Vim Script 0.01% Dockerfile 0.05% Java 0.71% HTML 0.12% Ruby 0.01% Batchfile 0.01% Go 0.30% CMake 1.42% Objective-C 0.07%

phd's Introduction

Update: I have finished my PhD! Please click above for my dissertation, which I'm delighted to say won the award for PhD Award for Best Disseration in Scotland 2019-2020. This repository is now in an archival state.

My PhD

A monolithic repository for (almost) everything I have done while at the University of Edinburgh. Living an #open life.

Publications

Chris Cummins, Zacharias V. Fisches, Tal Ben-Nun, Torsten Hoefler, Hugh Leather. "ProGraML: Graph-based Deep Learning for Program Optimization and Analysis". arXiv:2003.10536. [source code].
Chris Cummins, "Deep Learning for Compilers". PhD thesis. [source code]. Build command: $ bazel build //docs/thesis.
Chris Cummins, Pavlos Petoumenos, Alastair Murray, Hugh Leather. "Compiler Fuzzing through Deep Learning". ISSTA '18. [source code]. [pdf]. Build command: $ bazel build //docs/2018_07_issta.
Chris Cummins, Pavlos Petoumenos, Alastair Murray, Hugh Leather. "DeepSmith: Compiler Fuzzing through Deep Learning". ACACES '18. [source code]. [pdf]. Build command: $ bazel build //docs/2018_07_acaces.
Chris Cummins, Pavlos Petoumenos, Zheng Wang, Hugh Leather. "End-to-end Deep Learning of Optimization Heuristics". PACT '17. [source code]. [pdf]. Build command: $ bazel build //docs/2017_09_pact.
Chris Cummins, Pavlos Petoumenos, Zheng Wang, Hugh Leather. "Synthesizing Benchmarks for Predictive Modeling". CGO '17. [source code]. [pdf]. [acm]. Build command: $ bazel build //docs/2017_02_cgo.
Chris Cummins, Pavlos Petoumenos, Michel Steuwer, Hugh Leather. "Autotuning OpenCL Workgroup Sizes". ACACES '16. [source code]. Build command: $ bazel build //docs/2016_07_acaces.
Chris Cummins, Pavlos Petoumenos, Michel Steuwer, Hugh Leather. "Towards Collaborative Performance Tuning of Algorithmic Skeletons". HLPGPU '16, HiPEAC. [source code]. [pdf]. Build command: $ bazel build //docs/2016_01_hlpgpu.
Chris Cummins, Pavlos Petoumenos, Michel Steuwer, Hugh Leather. "Autotuning OpenCL Workgroup Size for Stencil Patterns". ADAPT '16, HiPEAC. [source code]. [pdf]. [arxiv]. Build command: $ bazel build //docs/2016_01_adapt.
Chris Cummins. "Autotuning Stencils Codes with Algorithmic Skeletons". MSc Thesis, 2015. The University of Edinburgh. [source code]. Build command: $ bazel build //docs/2015_08_msc_thesis.

Talks

Chris Cummins. "Compiler Fuzzing through Deep Learning", 3rd August, 2018. Codeplay, Edinburgh, Scotland. [files]. [slides].
Chris Cummins. "Machine Learning for Compilers", 20th July, 2018. Workshop on Introspective Systems for Automatically Generating Tests (ISAGT), Amsterdam, Netherlands. [files]. [pdf].
Chris Cummins. "Compiler Fuzzing through Deep Learning", 16th July, 2018. ACM SIGSOFT International Symposium on Software Testing and Analysis (ISSTA), Amsterdam, Netherlands. [files]. [slides].
Chris Cummins. "End-to-end Deep Learning of Optimization Heuristics", 23rd March, 2018. Facebook, Menlo Park. [files]. [slides].
Chris Cummins. "End-to-end Deep Learning of Optimization Heuristics", 2nd Feb, 2018. Google, Mountain View. [files]. [slides].
Chris Cummins. "Second Year Progression Review", 18th Dec, 2017. The University of Edinburgh, Scotland. [files]. [pdf].
Chris Cummins. "End-to-end Deep Learning of Optimization Heuristics", 4th Oct, 2017. The University of Edinburgh, Scotland. [files]. [slides].
Chris Cummins. "End-to-end Deep Learning of Optimization Heuristics", 12th Sep, 2017. International Conference on Parallel Architectures and Compilation Techniques (PACT), Portland, Oregon, USA. [files]. [slides].
Chris Cummins. "Deep Learning in Compilers", 14th Jun, 2017. The University of Edinburgh, Scotland. [files]. [pdf]. [transcript].
Chris Cummins. "Using Deep Learning to Generate Human-like Code", 22nd April, 2017. Scottish Programming Languages Seminar, University of St. Andrews, Scotland. [files]. [pdf].
Chris Cummins. "Synthesizing Benchmarks for Predictive Modeling", 6th Febuary, 2017. International Symposium on Code Generationand Optimization (CGO), Austin, Texas, USA. [files]. [slides].
Chris Cummins. "Machine Learning & Compilers", 9th September, 2016. Codeplay, Edinburgh, Scotland. [files]. [slides].
Chris Cummins. "Building an AI that Codes", 22nd July, 2016. Ocado Technology, Hatfield, England. [files]. [pdf].
Chris Cummins. "All the OpenCL on GitHub: Teaching an AI to code, one character at a time", 19th May, 2016. Amazon Development Centre, Edinburgh, Scotland. [files]. [pdf].
Chris Cummins. "Autotuning and Algorithmic Skeletons", Wed 10th Feb, 2016. The University of Edinburgh, Scotland. [files]. [pdf].
Chris Cummins. "Towards Collaborative Performance Tuning of Algorithmic Skeletons", Tues 19th Jan, 2016. HLPGPU, HiPEAC, Prague. [files]. [pdf].
Chris Cummins. "Autotuning OpenCL Workgroup Size for Stencil Patterns", Mon 18th Jan, 2016. ADAPT, HiPEAC, Prague. [files]. [pdf].
Chris Cummins. "Towards Collaborative Performance Tuning of Algorithmic Skeletons", Thurs 14th Jan, 2016. The University of Edinburgh, Scotland. [files]. [pdf].

Misc

Curriculum Vitae. [source code]. [pdf]. [html]. Build command: $ bazel build //docs/cv.
Chris Cummins, Pavlos Petoumenos, Michel Steuwer, Hugh Leather. "Collaborative Autotuning of Algorithmic Skeletons for GPUs and CPUs". Incomplete journal version of ADAPT and HLPGPU papers. [source code]. Build command: $ bazel build //docs/2016_12_wip_taco.
Chris Cummins. "Deep Learning for Compilers". PhD First Year Review Document, 2016. [source code]. Build command: $ bazel build //docs/2016_11_first_year_review.
Chris Cummins, Hugh Leather. "Autotuning OpenCL Workgroup Sizes". Rejected submission for PACT'16 Student Research Competition. [source code]. Build command: $ bazel build //docs/2016_07_pact.
Chris Cummins, Pavlos Petoumenos, Michel Steuwer, Hugh Leather. "Autotuning OpenCL Workgroup Sizes". Submission for PLDI'16 Student Poster Session. [source code]. Build command: $ bazel build //docs/2016_06_pldi.
Chris Cummins. "Autotuning and Skeleton-aware Compilation". PhD Progression Review, 2015. [source code]. Build command: $ bazel build //docs/2015_09_progression_review.

Building the code

See INSTALL.md and CONTRIBUTING.md.

phd's People

Contributors

Stargazers

Watchers

phd's Issues

//deeplearning/clgen default sample crashes in new docker container

Hi @ChrisCummins,

I've followed the instructions in the wiki for clgen in a new docker container, installed the dependencies and bazel.

bazel build //deeplearning/clgen compiles ok, but when I run bazel-out/*-py3-opt/bin/deeplearning/clgen/clgen --config deeplearning/clgen/test/data/tiny/config.pbtxt

I first get an encoding error

File "/phd/bazel-out/k8-py3-opt/bin/deeplearning/clgen/clgen.runfiles/phd/deeplearning/clgen/clgen.py", line 15
SyntaxError: Non-ASCII character '\xe2' in file /phd/bazel-out/k8-py3-opt/bin/deeplearning/clgen/clgen.runfiles/phd/deeplearning/clgen/clgen.py on line 16, but no encoding declared; see http://python.org/dev/peps/pep-0263/ for details

which I can correct by putting a # -*- coding: utf-8 -*- at the start of deeplearning/clgen/clgen.py but then I get errors at:

File "/phd/bazel-out/k8-py3-opt/bin/deeplearning/clgen/clgen.runfiles/phd/deeplearning/clgen/clgen.
py", line 96
    self.working_dir: pathlib.Path = pathlib.Path(
                    ^
SyntaxError: invalid syntax

Am I forgetting to set an environment variable, or am I missing something with this build process? Sorry, I've not used basel before, but thought you might be able to quickly point me in the right direction...

Beau.

Don't install python packages globally

Use virtualenvs

How to run the DeepSmith

Hi,
Thanks for your great work about fuzzing with DL,
Recently, I have read your paper and want to do some work after the DeepSmith.
But I have no idea about how to run your code.
Would you mind giving me a specific tutorial？
Thank you very much!

Defer evaluating `required=True` on labm8.app flags until main method execution

Currently, flags declared as required are checked during initial module evaluation. This is extremely awkward for things like unit tests files, as importing the module causes the flag presence to be checked.

One possible solution would be to evaluate whether a required flag is set directly before app.Run evaluates a user method.

Get an error on `bazel test //... `

Hi，Mr.Cummins
When I install the environment according to the README.md file in the phd directory, I execute bazel test //... The following error occurred,

sometimes the following error occurred,

Can you give me some advice on how to solve this problem？

How to set importer field?

When I wanted to extract individual source files from the cloned repos and import them into a contentfiles database, and run the code as following:
$ bazel run //datasets/github/scrape_repos:importer -- \ --clone_list $PWD/clone_list.pbtxt

I got an error said:
/datasets/github/scrape_repos/importer.py", line 136, in ImportFromLanguage raise ValueError('LanguageToClone.importer field not set') ValueError: LanguageToClone.importer field not set
My question is how to set the importer field.
Thank you.

Erros on 'bazel test //deeplearning/clgen/...'

Hi Chris,
I run the code 'bazel test //deeplearning/clgen/...' to test the environment after installation,
but get errors as follows:
'//deeplearning/clgen/models:keras_backend_test FAILED in 24.1s
/home/georgia/.cache/bazel/_bazel_georgia/03a359912ab05c7f13a1dd9a75789983/execroot/phd/bazel-out/k8-py3-opt/testlogs/deeplearning/clgen/models/keras_backend_test/test.log
//deeplearning/clgen/preprocessors:java_test FAILED in 1.6s
/home/georgia/.cache/bazel/_bazel_georgia/03a359912ab05c7f13a1dd9a75789983/execroot/phd/bazel-out/k8-py3-opt/testlogs/deeplearning/clgen/preprocessors/java_test/test.log
Executed 21 out of 21 tests: 19 tests pass and 2 fail locally.'

When I check the log,
I found some errors like
'AttributeError: 'Corpus' object has no attribute 'decoder'

deeplearning/clgen/models/keras_backend.py:92: AttributeError
'
and
'
File "/home/georgia/.cache/bazel/_bazel_georgia/03a359912ab05c7f13a1dd9a75789983/sandbox/linux-sandbox/289/execroot/phd/bazel-out/k8-py3-opt/bin/deeplearning/clgen/preprocessors/java_test.runfiles/phd/deeplearning/clgen/preprocessors/java.py", line 15, in
from lib.labm8 import bazelutil
ModuleNotFoundError: No module named 'lib.labm8'
'
Thank you.

Use DeepSmith to generate programs in other languages

Hi @ChrisCummins .

I'm trying to use your neural network to generate kernels in other languages. I'm currently following the comment #30 (comment). Is that the best way to do it?

I tried to create a corpus following this: https://github.com/ChrisCummins/phd/tree/master/datasets/github/scrape_repos.
All the commands ran, I scraped and cloned the repos, and when I run importer it gives me

Importing 0 $LANG repos ...

Is that right? Am I missing something?

Another question, given a tar.gz file with source code files, how do I create a CLgen model? How can I specify a config file for it? I tried follow this file, changing the work_dir, the local_tar_archive and the sampler , but when I run bazel run //deeplearning/clgen -- --config=/path/to/the/config/file, it gives me the error:

clgen.py:176] [Errno 2] No such file or directory: '/tmp/clgen_corpus_lh_2nyrr/corpus/some_file.c' (FileNotFoundError)

s/utc_epoch/unix_epoch/g

Rather embarrassingly, I have just learned that to this day I have been referring to unix epoch incorrectly as utc epoch. Not a high priority fix, but it's worth doing to save face. Before doing any automated refactoring, I'll need to check for text format protos which may contain a utc_epoch field, and sed them.

learn/algos: Implement Linear Regression

http://eli.thegreenplace.net/2016/linear-regression/

Split labm8 into repo and track as submodule

Add a third party javac package

The target //deeplearning/clgen/preprocessors:java implicitly depends on javac. I should make this dependency explicit rather than relying on system jdk packages.

me_db: Add a bokeh server to visualize stats

Split tools into make_tools and git_tools

Implement ustl::multiset

Port autotex to Makefile

Add Travis CI job

Remove extern/

If it's an external dependency then download a release tarball and build from that, else move it somewhere else.

learn/algos: Implement dimensionality reduction

http://colah.github.io/posts/2014-10-Visualizing-MNIST/

Add a //labm8/sh package for bash scripting

It should define useful functions like log, warning, fatal, etc.
Plus a system for creating scoped temporary directories (for smoke tests).
Bonus points if it can replicate gflags.

Refactor clgen.Corpus class to permit different types of corpus

Make clgen.Corpus abstract, and have concrete subclasses for the different types of corpuses: pre-encoded, local-files, etc.

Restructure 'labm8' to support multiple languages.

My codebase is increasingly polylingual, and I'm re-implementing some of the labm8 goodness in different languages. To save headaches, I should merge all my reusable library code into //labm8, and have subpackages for each language.

This large-scale refactoring consists of a couple of steps:

Move //labm8 into //labm8/py.
Move //phd into //labm8/cpp.
Merge //lib/phd into //labm8/cpp.
Make a release to github.

Build broken with new versions of bazel.

A handful of Skylark rules have been deprecated and moved in new bazel releases (> 0.18.1).

Update to new bazel python toolchains

Update BUILD files to use new python toolchains:

https://github.com/bazelbuild/rules_python/blob/master/proposals/2019-02-12-design-for-a-python-toolchain.md

This should remove the need for my ad-hoc symlinking of python in ./configure, and remove the need for .env.

me_db: Add fake dataset generator for HealthKit

Complete bazel build coverage

Current status:

Unhandled exception thrown during build; message: Unrecoverable error while evaluating node 'REPOSITORY_DIRECTORY:@org_pubref_rules_protobuf' (requested by nodes 'REPOSITORY:@org_pubref_rules_protobuf')

Hi，Mr.Cummins:
When I install the environment according to the README.md file in the phd directory, I execute bazel test //... or bazel test //lib/labm8:fs_test,error is as follows：

Starting local Bazel server and connecting to it...
...........
Unhandled exception thrown during build; message: Unrecoverable error while evaluating node 'REPOSITORY_DIRECTORY:@org_pubref_rules_protobuf' (requested by nodes 'REPOSITORY:@org_pubref_rules_protobuf')
INFO: Elapsed time: 0.639s
INFO: 0 processes.
FAILED: Build did NOT complete successfully (0 packages loaded)
java.lang.RuntimeException: Unrecoverable error while evaluating node 'REPOSITORY_DIRECTORY:@org_pubref_rules_protobuf' (requested by nodes 'REPOSITORY:@org_pubref_rules_protobuf')
at com.google.devtools.build.skyframe.AbstractParallelEvaluator$Evaluate.run(AbstractParallelEvaluator.java:441)
at com.google.devtools.build.lib.concurrent.AbstractQueueVisitor$WrappedRunnable.run(AbstractQueueVisitor.java:355)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
at java.lang.Thread.run(Thread.java:748)
Caused by: java.lang.IllegalArgumentException: Integer value http.postBuffer out of range
at org.eclipse.jgit.lib.DefaultTypedConfigGetter.getInt(DefaultTypedConfigGetter.java:153)
at org.eclipse.jgit.lib.Config.getInt(Config.java:263)
at org.eclipse.jgit.transport.HttpConfig.init(HttpConfig.java:229)
at org.eclipse.jgit.transport.HttpConfig.(HttpConfig.java:202)
at org.eclipse.jgit.transport.TransportHttp.(TransportHttp.java:280)
at org.eclipse.jgit.transport.TransportHttp$1.open(TransportHttp.java:207)
at org.eclipse.jgit.transport.Transport.open(Transport.java:570)
at org.eclipse.jgit.transport.Transport.open(Transport.java:446)
at org.eclipse.jgit.transport.Transport.open(Transport.java:325)
at org.eclipse.jgit.transport.Transport.open(Transport.java:294)
at org.eclipse.jgit.api.FetchCommand.call(FetchCommand.java:228)
at org.eclipse.jgit.api.CloneCommand.fetch(CloneCommand.java:306)
at org.eclipse.jgit.api.CloneCommand.call(CloneCommand.java:200)
at com.google.devtools.build.lib.bazel.repository.GitCloner.clone(GitCloner.java:199)
at com.google.devtools.build.lib.bazel.repository.GitRepositoryFunction.fetch(GitRepositoryFunction.java:73)
at com.google.devtools.build.lib.rules.repository.RepositoryDelegatorFunction.compute(RepositoryDelegatorFunction.java:188)
at com.google.devtools.build.skyframe.AbstractParallelEvaluator$Evaluate.run(AbstractParallelEvaluator.java:363)
... 4 more
java.lang.RuntimeException: Unrecoverable error while evaluating node 'REPOSITORY_DIRECTORY:@org_pubref_rules_protobuf' (requested by nodes 'REPOSITORY:@org_pubref_rules_protobuf')
at com.google.devtools.build.skyframe.AbstractParallelEvaluator$Evaluate.run(AbstractParallelEvaluator.java:441)
at com.google.devtools.build.lib.concurrent.AbstractQueueVisitor$WrappedRunnable.run(AbstractQueueVisitor.java:355)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
at java.lang.Thread.run(Thread.java:748)
Caused by: java.lang.IllegalArgumentException: Integer value http.postBuffer out of range
at org.eclipse.jgit.lib.DefaultTypedConfigGetter.getInt(DefaultTypedConfigGetter.java:153)
at org.eclipse.jgit.lib.Config.getInt(Config.java:263)
at org.eclipse.jgit.transport.HttpConfig.init(HttpConfig.java:229)
at org.eclipse.jgit.transport.HttpConfig.(HttpConfig.java:202)
at org.eclipse.jgit.transport.TransportHttp.(TransportHttp.java:280)
at org.eclipse.jgit.transport.TransportHttp$1.open(TransportHttp.java:207)
at org.eclipse.jgit.transport.Transport.open(Transport.java:570)
at org.eclipse.jgit.transport.Transport.open(Transport.java:446)
at org.eclipse.jgit.transport.Transport.open(Transport.java:325)
at org.eclipse.jgit.transport.Transport.open(Transport.java:294)
at org.eclipse.jgit.api.FetchCommand.call(FetchCommand.java:228)
at org.eclipse.jgit.api.CloneCommand.fetch(CloneCommand.java:306)
at org.eclipse.jgit.api.CloneCommand.call(CloneCommand.java:200)
at com.google.devtools.build.lib.bazel.repository.GitCloner.clone(GitCloner.java:199)
at com.google.devtools.build.lib.bazel.repository.GitRepositoryFunction.fetch(GitRepositoryFunction.java:73)
at com.google.devtools.build.lib.rules.repository.RepositoryDelegatorFunction.compute(RepositoryDelegatorFunction.java:188)
at com.google.devtools.build.skyframe.AbstractParallelEvaluator$Evaluate.run(AbstractParallelEvaluator.java:363)
FAILED: Build did NOT complete successfully (0 packages loaded)

Could you give me some help? Thank you!

Add a tutorial for porting DL program generation to other languages

In the paper, you mentioned that your RNN can easily be ported to any other programming language. We are trying to validate that claim . Can you provide tutorial if it is possible to just use the RNN model in the code base?

Split rules into make_tools/rules/<lang>.make files

lib.labm8.lockfile.UnableToAcquireLockError

Hi Chris,
When I run the command "bazel run //docs/2018_07_issta/artifact_evaluation:01_evaluate_generator", got a error as follows:

File "/home/gxy/.cache/bazel/_bazel_gxy/45b7240206d651f2bfc82ad3534122dd/execroot/phd/bazel-out/local-opt/bin/docs/2018_07_issta/artifact_evaluation/01_evaluate_generator.runfiles/phd/lib/labm8/lockfile.py", line 173, in acquire
raise UnableToAcquireLockError(self)
lib.labm8.lockfile.UnableToAcquireLockError: Unable to acquire file lock owned by a different process.
Lock acquired by process 6048 on gxy@localhost at 2018-07-30 05:56:28.414000.
Lock path: /tmp/phd/docs/2018_07_issta/artifact_evaluation/clgen/model/15b5ca3baf7cae9160f1f5ff3f47f8ecfe279098/LOCK
ERROR: Non-zero return code '1' from command: Process exited with status 1.

PS：The operating system is Centos 7.5.

How to solve it. Thanks

Split targets for external libs into make_tools/targets/<lib>.make files

Remove til/ dir

Put it into offline reference files

Build google benchmark from release tarball

Remove //lib directory.

There are two issues with //lib:

a python module shadowing issue that prevents importing lib.labm8 (e.g. #33).
the term lib is too vague to be useful. lots of my code could be considered library code, and some of the code currently in lib isn't.

I propose migrating the current residents of the lib dir as follows:

//lib/dack -> //system/dpack.
//lib/labm8 -> //labm8.
//lib/phd to be deleted, superseded by //phd.
//lib/stl -> //learn/stl.
//lib/tex -> //docs/libtex.

freefocus: Complete prototype of RPC service

Replace ad-hoc pre-commit hooks with pre-commit project

All instances of python subprocesses should have a timeout set

Refactor all subprocess.Popen(), subprocess.check_output(), subprocess.check_call() etc to ensure that they timeout -s9 <timeout_seconds>.

//datasets/github/scrape_repos:scraper_test has undeclared dependency.

We shouldn't hardcode the ~/.githubrc file.

Add an (optional) makeignore.txt file

e.g.

$ cat ~/phd/makeignore.txt
projects
playground/rt

Will cause projects/ and playground/rt files to be skipped.

Complete conversion to new repo style

Projects in need of tests

//system/dotfiles
//util/shutterbug
//util/photolint
//util/lmk

Projects in need of refactoring to new style

//deeplearning/clgen
//gpu/cldrive

learn/algos: Implement string similarity

https://github.com/tdebatty/java-string-similarity

me_db: Investigate lightroom metadata import

No bootstrap.sh

When I run the install.sh in the directory: phd/docs/2018_07_issta/artifact_evaluation, I got the following error:

+++ dirname ./install.sh
++ cd ./../../..
++ pwd

ROOT=/home/gxye/project/phd
main
cd /home/gxye/project/phd
test -f /home/gxye/project/phd/.env
/home/gxye/project/phd/tools/bootstrap.sh
./install.sh: line 23: /home/gxye/project/phd/tools/bootstrap.sh: No such file or directory

I checked the target directory "./phd/tools/", and there is not bootstrap.sh, only the bootstrap.py.

Got An Error on ‘docker run chriscummins/opencl_fuzz’

Hi，Mr.Cummins
When I try to do docker run chriscummins/opencl_fuzz I got an error like:
I0728 03:12:19.383133 140557201053440 opencl_fuzz.py:331] Recording interesting results in /out. I0728 03:12:19.383535 140557201053440 opencl_fuzz.py:333] Preparing generator. Traceback (most recent call last): File "/app/experimental/deeplearning/deepsmith/opencl_fuzz/opencl_fuzz_image.binary.runfiles/phd/experimental/deeplearning/deepsmith/opencl_fuzz/opencl_fuzz.py", line 373, in <module> app.run(main) File "/app/experimental/deeplearning/deepsmith/opencl_fuzz/opencl_fuzz_image.binary.runfiles/pypi__absl_py_0_1_10/absl/app.py", line 274, in run _run_main(main, argv) File "/app/experimental/deeplearning/deepsmith/opencl_fuzz/opencl_fuzz_image.binary.runfiles/pypi__absl_py_0_1_10/absl/app.py", line 238, in _run_main sys.exit(main(argv)) File "/app/experimental/deeplearning/deepsmith/opencl_fuzz/opencl_fuzz_image.binary.runfiles/phd/experimental/deeplearning/deepsmith/opencl_fuzz/opencl_fuzz.py", line 336, in main clgen_pretrained.ClgenGenerator) File "/app/experimental/deeplearning/deepsmith/opencl_fuzz/opencl_fuzz_image.binary.runfiles/phd/experimental/deeplearning/deepsmith/opencl_fuzz/opencl_fuzz.py", line 311, in GeneratorFromFlag return generator_class(config) File "/app/experimental/deeplearning/deepsmith/opencl_fuzz/opencl_fuzz_image.binary.runfiles/phd/deeplearning/deepsmith/generators/clgen_pretrained.py", line 53, in __init__ self.instance = sample.Instance(self.config.instance) File "/app/experimental/deeplearning/deepsmith/opencl_fuzz/opencl_fuzz_image.binary.runfiles/phd/deeplearning/clgen/sample.py", line 46, in __init__ pathlib.Path(config.pretrained_model)) File "/app/experimental/deeplearning/deepsmith/opencl_fuzz/opencl_fuzz_image.binary.runfiles/phd/deeplearning/clgen/models/pretrained.py", line 31, in __init__ self.cache = cache.FSCache(self.path) AttributeError: module 'deeplearning.clgen.cache' has no attribute 'FSCache'

How can I fix this?

mkdir: created directory '/phd/docs/2018-07-issta/artifact_evaluation/03_evaluate_results/output'
Traceback (most recent call last):
  File "/phd/docs/2018-07-issta/artifact_evaluation/03_evaluate_results/experiments.py", line 20, in <module>
    from deeplearning.deepsmith import datastore
  File "/phd/deeplearning/deepsmith/datastore.py", line 10, in <module>
    import deeplearning.deepsmith.client
  File "/phd/deeplearning/deepsmith/client.py", line 2, in <module>
    from deeplearning.deepsmith import db
  File "/phd/deeplearning/deepsmith/db.py", line 12, in <module>
    from deeplearning.deepsmith.proto import datastore_pb2
  File "/phd/deeplearning/deepsmith/proto/datastore_pb2.py", line 25, in <module>
    dependencies=[deeplearning_dot_deepsmith_dot_proto_dot_deepsmith__pb2.DESCRIPTOR,])
  File "/phd/venv/phd/lib/python3.6/site-packages/google/protobuf/descriptor.py", line 829, in __new__
    return _message.default_pool.AddSerializedFile(serialized_pb)
TypeError: Couldn't build proto file into descriptor pool!
Invalid proto descriptor for file "datastore.proto":
  datastore.proto: Import "deeplearning/deepsmith/proto/deepsmith.proto" has not been loaded.
  deepsmith.SubmitTestcasesRequest.testcases: "deepsmith.Testcase" seems to be defined in "deepsmith.proto", which is not imported by "datastore.proto".  To use it here, please add the necessary import.
  deepsmith.RequestTestcasesRequest.testbed: "deepsmith.Testbed" seems to be defined in "deepsmith.proto", which is not imported by "datastore.proto".  To use it here, please add the necessary import.
  deepsmith.RequestTestcasesRequest.harness: "deepsmith.Harness" seems to be defined in "deepsmith.proto", which is not imported by "datastore.proto".  To use it here, please add the necessary import.
  deepsmith.RequestTestcasesResponse.testcases: "deepsmith.Testcase" seems to be defined in "deepsmith.proto", which is not imported by "datastore.proto".  To use it here, please add the necessary import.
  deepsmith.SubmitResultsRequest.results: "deepsmith.Result" seems to be defined in "deepsmith.proto", which is not imported by "datastore.proto".  To use it here, please add the necessary import.
  deepsmith.ResultsCollection.results: "deepsmith.Result" seems to be defined in "deepsmith.proto", which is not imported by "datastore.proto".  To use it here, please add the necessary import.
  deepsmith.ResultsTable.ResultsTableEntry.testbed: "deepsmith.Testbed" seems to be defined in "deepsmith.proto", which is not imported by "datastore.proto".  To use it here, please add the necessary import.
  deepsmith.ResultsTable.ResultsTableEntry.harness: "deepsmith.Harness" seems to be defined in "deepsmith.proto", which is not imported by "datastore.proto".  To use it here, please add the necessary import.
  deepsmith.ResultsTable.ResultsTableEntry.results: "deepsmith.Result" seems to be defined in "deepsmith.proto", which is not imported by "datastore.proto".  To use it here, please add the necessary import.

I don't think I know how to solve this one by myself any ideas what might be wrong?