We have submitted a submission.zip containing:
- metadata
- model.py
- SubmissionModel
- - assets
- - variables
- - keras_metadata.pb
- - saved_model.pb
We have noticed that the folder assets is empty, may it be a problem?
metadata and model.py are the ones provided in starting_kit
The content of SubmissionModel is exactly the output of saving process on Colab
The log of the error is:
WARNING: Your kernel does not support swap limit capabilities or the cgroup is not mounted. Memory limited without swap.
Traceback (most recent call last):
File "/multiverse/storage/lattari/Prj/phd/AN2DL/Competition_1_running_dir/worker_gpu0_dir/tmp/codalab/tmpwgaoSd/run/program/score.py", line 106, in
from model import model # example model, in scikit-learn style
File "/multiverse/storage/lattari/Prj/phd/AN2DL/Competition_1_running_dir/worker_gpu0_dir/tmp/codalab/tmpwgaoSd/run/input/res/model.py", line 1, in
{"nbformat":4,"nbformat_minor":0,"metadata":{"colab":{"name":"model.py","provenance":[],"authorship_tag":"ABX9TyOgHjW9nDDQ/EruHNAATcGV"},"kernelspec":{"name":"python3","display_name":"Python 3"},"language_info":{"name":"python"}},"cells":[{"cell_type":"code","metadata":{"id":"dcI0B26R6y4R"},"source":["import os\n","import tensorflow as tf\n","\n","class model:\n"," def init(self, path):\n"," self.model = tf.keras.models.load_model(os.path.join(path, 'SubmissionModel'))\n","\n"," def predict(self, X):\n","\n"," out = self.model.predict(X)\n"," out = tf.argmax(out, axis=-1)\n","\n"," return out"],"execution_count":null,"outputs":[]}]}
NameError: name 'null' is not defined
Please follow exactly the instructions we provided. Double check the submission.zip file, it is not structured as described in your topic.
Posted by: an2dl.competitions @ Nov. 15, 2021, 10:39 a.m.