search

antoine77340 / Youtube-8M-WILLOW

Kaggle Youtube 8M WILLOW approach
Apache License 2.0
465 stars 165 forks source link

Request your trained model #7

Closed feiyun1265 closed 6 years ago

feiyun1265 commented 6 years ago

Congratulations on your winning. And thanks for sharing your work. I'm very interested in the YouTube-8M Large-Scale Video Understanding project. Now, i want to use my own video test your team model. Can you send me your trained model. My email:chshfmail@163.com. Thanks.

antoine77340 commented 6 years ago

Hi, Yes sure, I just realized Google had released their code for feature extraction. I will update the github repo with the pretrained best model I have (If I find it back ahha :) ).

antoine77340 commented 6 years ago

Here it is: https://www.rocq.inria.fr/cluster-willow/amiech/pretrainedmodel.zip This is the model for : python inference.py --output_file=test-gatednetvladLF-256k-1024-80-0002-300iter-norelu-basic-gatedmoe.csv --input_data_pattern="$path_to_features/test*.tfrecord" --model=NetVLADModelLF --train_dir=gatednetvladLF-256k-1024-80-0002-300iter-norelu-basic-gatedmoe --frame_features=True --feature_names="rgb,audio" --feature_sizes="1024,128" --batch_size=1024 --base_learning_rate=0.0002 --netvlad_cluster_size=256 --netvlad_hidden_size=1024 --moe_l2=1e-6 --iterations=300 --learning_rate_decay=0.8 --netvlad_relu=False --gating=True --moe_prob_gating=True --run_once=True --top_k=50

feiyun1265 commented 6 years ago

Thanks for your sharing. @antoine77340