{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:PQ3ETXENEYJHHE2U2FAJLFUYRN","short_pith_number":"pith:PQ3ETXEN","schema_version":"1.0","canonical_sha256":"7c3649dc8d2612739354d1409596988b7e86f76d021e364feacd0ff796ec3ec2","source":{"kind":"arxiv","id":"2005.02130","version":1},"attestation_state":"computed","paper":{"title":"Importance of Data Loading Pipeline in Training Deep Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Mahdi Zolnouri, Vahid Partovi Nia, Xinlin Li","submitted_at":"2020-04-21T14:19:48Z","abstract_excerpt":"Training large-scale deep neural networks is a long, time-consuming operation, often requiring many GPUs to accelerate. In large models, the time spent loading data takes a significant portion of model training time. As GPU servers are typically expensive, tricks that can save training time are valuable.Slow training is observed especially on real-world applications where exhaustive data augmentation operations are required. Data augmentation techniques include: padding, rotation, adding noise, down sampling, up sampling, etc. These additional operations increase the need to build an efficient"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2005.02130","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2020-04-21T14:19:48Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"ea92dc29cbcd50b2b034c40ed3fdc4e9b4879ae8b92e1ac68371e699c062d490","abstract_canon_sha256":"a8440ca3ed9efe2740aefad6c31fde62761bb30b1f3df42e217b3381f33bab9b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:00:22.854010Z","signature_b64":"Cf/NIm2oCvTSZQ1BA6rcO5HposqukkO+Pls55XA/LYHxwWKCZHJs8NArkglMYRwc/51WfdqESoQb9in2fLaDCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7c3649dc8d2612739354d1409596988b7e86f76d021e364feacd0ff796ec3ec2","last_reissued_at":"2026-07-05T01:00:22.853538Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:00:22.853538Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Importance of Data Loading Pipeline in Training Deep Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Mahdi Zolnouri, Vahid Partovi Nia, Xinlin Li","submitted_at":"2020-04-21T14:19:48Z","abstract_excerpt":"Training large-scale deep neural networks is a long, time-consuming operation, often requiring many GPUs to accelerate. In large models, the time spent loading data takes a significant portion of model training time. As GPU servers are typically expensive, tricks that can save training time are valuable.Slow training is observed especially on real-world applications where exhaustive data augmentation operations are required. Data augmentation techniques include: padding, rotation, adding noise, down sampling, up sampling, etc. These additional operations increase the need to build an efficient"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2005.02130","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2005.02130/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2005.02130","created_at":"2026-07-05T01:00:22.853603+00:00"},{"alias_kind":"arxiv_version","alias_value":"2005.02130v1","created_at":"2026-07-05T01:00:22.853603+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2005.02130","created_at":"2026-07-05T01:00:22.853603+00:00"},{"alias_kind":"pith_short_12","alias_value":"PQ3ETXENEYJH","created_at":"2026-07-05T01:00:22.853603+00:00"},{"alias_kind":"pith_short_16","alias_value":"PQ3ETXENEYJHHE2U","created_at":"2026-07-05T01:00:22.853603+00:00"},{"alias_kind":"pith_short_8","alias_value":"PQ3ETXEN","created_at":"2026-07-05T01:00:22.853603+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2411.16618","citing_title":"StructFormer: Document Structure-based Masked Attention and its Impact on Language Model Pre-Training","ref_index":21,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PQ3ETXENEYJHHE2U2FAJLFUYRN","json":"https://pith.science/pith/PQ3ETXENEYJHHE2U2FAJLFUYRN.json","graph_json":"https://pith.science/api/pith-number/PQ3ETXENEYJHHE2U2FAJLFUYRN/graph.json","events_json":"https://pith.science/api/pith-number/PQ3ETXENEYJHHE2U2FAJLFUYRN/events.json","paper":"https://pith.science/paper/PQ3ETXEN"},"agent_actions":{"view_html":"https://pith.science/pith/PQ3ETXENEYJHHE2U2FAJLFUYRN","download_json":"https://pith.science/pith/PQ3ETXENEYJHHE2U2FAJLFUYRN.json","view_paper":"https://pith.science/paper/PQ3ETXEN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2005.02130&json=true","fetch_graph":"https://pith.science/api/pith-number/PQ3ETXENEYJHHE2U2FAJLFUYRN/graph.json","fetch_events":"https://pith.science/api/pith-number/PQ3ETXENEYJHHE2U2FAJLFUYRN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PQ3ETXENEYJHHE2U2FAJLFUYRN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PQ3ETXENEYJHHE2U2FAJLFUYRN/action/storage_attestation","attest_author":"https://pith.science/pith/PQ3ETXENEYJHHE2U2FAJLFUYRN/action/author_attestation","sign_citation":"https://pith.science/pith/PQ3ETXENEYJHHE2U2FAJLFUYRN/action/citation_signature","submit_replication":"https://pith.science/pith/PQ3ETXENEYJHHE2U2FAJLFUYRN/action/replication_record"}},"created_at":"2026-07-05T01:00:22.853603+00:00","updated_at":"2026-07-05T01:00:22.853603+00:00"}