Things went wrong in the debug log here:
drwxr-sr-x 2 root aicrowd 4096 May 11 04:03 utilities
Traceback (most recent call last):
File "run.py", line 1, in <module>
from client_launcher import start_test_client
File "/src/client_launcher.py", line 6, in <module>
from aicrowd_wrapper import AIcrowdWrapper
File "/src/aicrowd_wrapper.py", line 6, in <module>
from models.user_config import UserModel
File "/src/models/user_config.py", line 13, in <module>
from models.Track1Model import Track1Model
ModuleNotFoundError: No module named 'models.Track1Model'
And this is my submission:
What did I do wrongly?
My Track1Model is in the models folder, and yet the inference engine cannot detect the folder. Do I need to adjust the Dockerfile, or do I need to move my model file elsewhere?
I think it should be an issue with your code. You should be able to locally debug that with the docker_run.sh.
This is my user_config.py
# Importing DummyModel from the models package.
# The DummyModel class is located in the dummy_model.py file inside the 'models' directory.
# from models.dummy_model import DummyModel
# This line establishes an alias for the DummyModel class to be used within this script.
# Instead of directly using DummyModel everywhere in the code, we're assigning it to 'UserModel'.
# This approach allows for easier reference to your model class when evaluating your models,
# UserModel = DummyModel
# When implementing your own model please follow this pattern:
#
from models.track1_model import Track1Model
#
# Replace 'your_model' with the name of your Python file containing the model class
# and 'YourModel' with the class name of your model.
#
# Finally, assign YourModel to UserModel as shown below to use it throughout your script.
#
UserModel = Track1Model
And I get this from my docker_run.sh terminal:
aicrowd/amazon-kddcup24-submission:493c8e4
2024/05/11 15:26:51 http2: server: error reading preface from client //./pipe/docker_engine: file has already been closed
2024/05/11 15:26:51 http2: server: error reading preface from client //./pipe/docker_engine: file has already been closed
#0 building with "default" instance using docker driver
#1 [internal] load build definition from Dockerfile
#1 transferring dockerfile: 1.54kB done
#1 DONE 0.0s
#2 [internal] load metadata for docker.io/nvidia/cuda:11.8.0-cudnn8-runtime-ubuntu20.04
#2 DONE 1.9s
#3 [internal] load .dockerignore
#3 transferring context: 49B done
#3 DONE 0.0s
#4 [ 1/10] FROM docker.io/nvidia/cuda:11.8.0-cudnn8-runtime-ubuntu20.04@sha256:c2336dadc71ae5b5ce490a55cc0100d876287d28f429a5d2840c8a3a8e86fef0
#4 DONE 0.0s
#5 [internal] load build context
#5 transferring context: 103B done
#5 DONE 0.0s
#6 [ 9/10] COPY --chown=1001:1001 requirements_eval.txt /home/aicrowd/requirements_eval.txt
#6 CACHED
#7 [ 4/10] RUN groupadd -g 1001 aicrowd && useradd -m -s /bin/bash -u 1001 -g aicrowd -G sudo aicrowd
#7 CACHED
#8 [ 5/10] WORKDIR /home/aicrowd
#8 CACHED
#9 [ 2/10] COPY apt.txt /tmp/apt.txt
#9 CACHED
#10 [ 7/10] COPY --chown=1001:1001 requirements.txt /home/aicrowd/requirements.txt
#10 CACHED
#11 [ 6/10] RUN wget -nv -O miniconda.sh https://repo.anaconda.com/miniconda/Miniconda3-py38_22.11.1-1-Linux-x86_64.sh && bash miniconda.sh -b -p /home/aicrowd/.conda && . /home/aicrowd/.conda/etc/profile.d/conda.sh && conda install cmake -y && conda clean -y -a && rm -rf miniconda.sh
#11 CACHED
#12 [ 8/10] RUN pip install -r requirements.txt --no-cache-dir
#12 CACHED
#13 [ 3/10] RUN apt -qq update && apt -qq install -y --no-install-recommends `cat /tmp/apt.txt | tr -d '\r'` locales wget build-essential && locale-gen en_US.UTF-8 && rm -rf /var/cache/apt/* /var/lib/apt/lists/* && apt clean
#13 CACHED
#14 [10/10] RUN pip install -r requirements_eval.txt --no-cache-dir
#14 CACHED
#15 exporting to image
#15 exporting layers done
#15 writing image sha256:ddca3a44f91918fdfb9a3cd187e48030ae71c735bc43c3951392cd8dfcfaa673 done
#15 naming to docker.io/aicrowd/amazon-kddcup24-submission:493c8e4 done
#15 DONE 0.0s
And when I tested it from local_evaluation.py, the program could test my model nice and well.
I did not make any changes to the Dockerfile ye. Should I have?
Please tag us under the issue of your submission. With the current information I cannot locate your submission.