spotify_mpd_two_tower
Building two tower models with Spotify data in GCP
File Explorer
- spellcheck.yml
- spelling_action.yml
- wordlist.txt
- 00-bq-data-prep-checkpoint.ipynb
- 01-bq-data-prep-checkpoint.ipynb
- 01-tfrecord-beam-pipeline-checkpoint.ipynb
- 01-z-deploy-query-model-CPR-checkpoint-checkpoint.ipynb
- 01_BQ_data_prep-checkpoint.ipynb
- 02-build-model-checkpoint.ipynb
- 02_TFRecord-checkpoint.ipynb
- 03-create-test-instances-checkpoint.ipynb
- 03-matching-engine-checkpoint.ipynb
- 03_built-in-two-tower-checkpoint.ipynb
- 04-create-test-instances-checkpoint.ipynb
- [JT]-00-bq-data-prep-checkpoint.ipynb
- [JT]-01-tfrecord-beam-pipeline-checkpoint.ipynb
- candidate_embeddings-checkpoint.json
- Parse Data-checkpoint.ipynb
- README-checkpoint.md
- predictor.cpython-37.pyc
- main-checkpoint.py
- README-checkpoint.md
- requirements-checkpoint.txt
- setup-checkpoint.py
- __init__.cpython-37.pyc
- __init__-checkpoint.py
- candidate_pipeline-checkpoint.py
- __init__.cpython-37.pyc
- candidate_pipeline.cpython-37.pyc
- __init__.py
- candidate_pipeline.py
- __init__.py
- main.py
- README.md
- requirements.txt
- setup.py
- main-train-checkpoint.py
- README-checkpoint.MD
- setup-checkpoint.py
- train_pipe_shape-checkpoint.py
- __init__.cpython-37.pyc
- train_pipe.cpython-37.pyc
- __init__.py
- train_pipe_shape.py
- __init__.py
- main-train.py
- README.MD
- setup.py
- vectorQuant-checkpoint.gif
- 2tower-feature-processing-f12.jpg
- activate-cloud-shell.png
- algo.png
- arch.png
- bq-schema-load-json-raw-f8.jpg
- candidate-track-from-struct-f10.jpg
- compute_sa.png
- create-a-tb.png
- create_keys.png
- deployed_endpoint.png
- embeddings.png
- experiment-console.png
- factorization-based-retrieval.jpg
- feat-engineering-decision-tree-f13.jpg
- feature-enrich-pipeline.png
- Loss_Types.max-1000x1000.png
- me-indexes.png
- me-resources.png
- me-resources2.png
- model-registry.png
- mpd-loaded-tracks-struct--f9.jpg
- ndr-retrieval.jpg
- new-notebook.png
- new-project.png
- new-terminal.png
- nvtop-optimized.png
- nvtop.png
- pl-fan-out-f11.jpg
- playlists-nested.png
- saved-models.png
- semantic-pair.png
- speedvsaccuracy.max-1600x1600.png
- spotify-dev-console.png
- subnetwork.png
- subnetwork2.png
- tb-loss.png
- tb-metrics.png
- tb-profiler.png
- tensorboard.png
- terminal.png
- tfrs-train-profiler-v1.png
- token-based-retrieval.jpg
- tracks-string.png
- train-dataset-metadata.png
- train-pipeline-sp-e2e.png
- unique-artists.png
- unique-songs.png
- upload-tb-logs.png
- vectorQuant.gif
- 01-evolution-of-retrieval.md
- 02-problem-framing.md
- 03-GCP-data-preprocessing-and-prep.md
- 04-feature-engineering.md
- 05-two-tower-encoders.md
- call_spotify_api_artist.py
- call_spotify_api_audio.py
- feature_store_configs.py
- __init__.py
- build_audio_ranker.py
- feature_sets.py
- requirements.txt
- task.py
- tf_ranking_model.py
- train_config.py
- train_utils.py
- create_ann_index.py
- create_ann_index_endpoint_vpc.py
- create_brute_force_index.py
- create_brute_index_endpoint_vpc.py
- create_tensorboard.py
- deploy_ann_index.py
- deploy_brute_index.py
- generate_candidates.py
- model_monitoring_config.py
- pipeline_config.py
- send_skewed_traffic.py
- test_model_endpoint.py
- test_model_index_endpoint.py
- train_custom_model.py
- __init__.py
- feature_sets.py
- interactive_train.py
- requirements.txt
- task.py
- test_instances.py
- train_config.py
- train_utils.py
- two_tower.py
- adapt_fixed_text_layer_vocab.py
- adapt_ragged_text_layer_vocab.py
- config.py
- create_master_vocab.py
- cloudbuild.yaml
- Dockerfile_rank
- Dockerfile_tfrs
- tb_profiling-checkpoint.ipynb
- accelerators.py
- feature_set_utils.py
- install-nvtop.sh
- local_utils.py
- tb_profiling.ipynb
- test_instances.py
- train_utils.py
- two_tower_16_bit_edition.py
- workerpool_specs.py
- [JT]-intro-ranker.ipynb
- ranking.md
- [JT]-02-build-lite-model.ipynb
- [JT]-XX-WIP-tfrs-vocab-pipe.ipynb
- [WIP]-vocab-pipe.ipynb
- check-vertex-beta-sdk.ipynb
- check-vocabs.ipynb
- data-pipeline.py
- two_tower_lite.py
- vocab_dict_old.pkl
- .DS_Store
- .gcloudignore
- .gitignore
- 00-env-setup.ipynb
- 00b-load-core-data-to-bq.ipynb
- 01-bq-data-prep.ipynb
- 02-tfrecord-beam-pipeline.ipynb
- 03-build-retrieval-model.ipynb
- 03b-build-retrieval-image.ipynb
- 04-custom-train-retrieval.ipynb
- 05-candidate-generation.ipynb
- 06-deploy-query-tower-monitoring.ipynb
- 06b-centralized-model-montioring-deploy-query-tower-monitoring.ipynb
- 07-train-pipeline.ipynb
- 08-build-ranking-model.ipynb
- 09b-build-rank-image.ipynb
- 09c-custom-train-rank.ipynb
- 10-feature-store.ipynb
- bonus-recs-for-your-spotify.ipynb
- clean-up-ME-indexes.ipynb
- custom_pipeline_spec.json
- pip_freeze.txt
- prediction_featurestore_fetch_config.yaml
- README.md
- train_job_dict.pkl
- vocab_dict.pkl
# Use via CDN
jsDelivrjsDelivr serves any public GitHub repository as a CDN with zero setup. Pick a version and a file to get a ready-to-paste link and snippet.
Command Glossary
Commands referenced in this DOCs, explained below.
gcloud projects
View Details ▼
gcloud projects
Manage project access policies in Google Cloud.
See also: `gcloud`.
gcloud projects create {{project_id|project_number}}
Create a new project:
gcloud projects list
List all active projects:
gcloud projects describe {{project_id}}
Display metadata for a project:
gcloud
View Details ▼
gcloud
The official CLI tool for Google Cloud Platform.
Some subcommands such as `app` and `init` have their own usage documentation.
gcloud config list
List all properties in one's active configuration:
gcloud auth login
Login to a Google account:
gcloud config set project {{project_name}}
Set the active project:
git clone
View Details ▼
git clone
Clone an existing repository.
git clone {{remote_repository_location}} {{path/to/directory}}
Clone an existing repository into a new directory (the default directory is the repository name):
git clone --recursive {{remote_repository_location}}
Clone an existing repository and its submodules:
git clone {{[-n|--no-checkout]}} {{remote_repository_location}}
Clone only the `.git` directory of an existing repository:
pop
View Details ▼
pop
Send emails from your terminal.
pop
Launch the Text-based User Interface:
pop < {{path/to/message.md}} --from {{[email protected]}} --to {{[email protected]}} --subject "{{On the Subject of Ducks...}}" --attach {{path/to/attachment}}
Send an email using the content of a Markdown file as body:
pop --help
Display help:
nvtop
View Details ▼
nvtop
Interactive ncurses-based GPU process and status viewer for AMD, Intel, and NVIDIA GPUs.
See also: `amdgpu_top`, `radeontop`.
nvtop
Launch the interactive GPU monitor:
nvtop {{[-d|--delay]}} {{2}}
Set the update delay in tenths of a second (for example, 2 = 0.2 seconds):
nvtop {{[-C|--no-color]}}
Run in monochrome (no color) mode:
