NLP-AICity2021

We have two codebases. For the final submission, we conduct the feature ensemble, where features are from two codebases.

1 – Merge all packages into one:

$ zip -FF AIC21_Track5_NL_Retrieval.zip --out AIC21_Track5_NL_Retrieval_full.zip

2 – Delete the smaller packages to save disk space (optional):

$ rm AIC21_Track5_NL_Retrieval.z*

3 – Unzip the merged package:

$ mkdir AIC21_Track5_NL_Retrieval
$ cd AIC21_Track5_NL_Retrieval
$ unzip -FF ../AIC21_Track5_NL_Retrieval_full.zip

python data/extract_vdo_frms.py

Name		Name	Last commit message	Last commit date
Latest commit History 89 Commits
checkpoints		checkpoints
doc		doc
tool		tool
.gitignore		.gitignore
README.md		README.md
__init__.py		__init__.py
auto_augment.py		auto_augment.py
baseline-results.json		baseline-results.json
circle_loss.py		circle_loss.py
config.py		config.py
default.yaml		default.yaml
evaluate.py		evaluate.py
extract_frame.sh		extract_frame.sh
extract_vdo_frms.py		extract_vdo_frms.py
generate_clear_query.py		generate_clear_query.py
generate_clear_train.py		generate_clear_train.py
model.py		model.py
prepare_motions.py		prepare_motions.py
prepare_output_json.py		prepare_output_json.py
random_erasing.py		random_erasing.py
requirements.txt		requirements.txt
resnet_t2v.py		resnet_t2v.py
results-clear.json		results-clear.json
results.json		results.json
results_hub.json		results_hub.json
run.sh		run.sh
sam.py		sam.py
siamese_baseline_model.py		siamese_baseline_model.py
swa_utils.py		swa_utils.py
test-re.py		test-re.py
test.py		test.py
train.bak		train.bak
train.py		train.py
train_xh.py		train_xh.py
tsm_util.py		tsm_util.py
utils.py		utils.py
utils_T2.py		utils_T2.py
val.py		val.py
vehicle_retrieval_dataset.py		vehicle_retrieval_dataset.py
wget-log		wget-log

Provide feedback