Upload Corpus-30M/config.yaml with huggingface_hub

Corpus-30M/config.yaml

@@ -38,81 +38,12 @@ model:
   per_view_normalization: false
   values_only_sanity_check: false
   data_loading_speed_sanity_check: false
-wandb:
-  entity: theislab-transformer
-  run_id: null
-  enabled: true
-  project: contrastive-transformer
-  run_name: cont/PE/padding_mask_True/flash_True/PDR_0.5/B512/P400-P60000/mixed-0.25/panel-.*
 datamodule:
   name: GeneToken
   columns:
   - cell_type
   - donor_id
   dataset:
-    val:
-      same:
-        max_tokens: 1000
-        variable_size: false
-        panel_size_max: 60000
-        panel_size_min: 400
-        panel_selection: mixed
-        panel_filter_regex: .*
-        panel_selection_mixed_prob: 0.25
-      within_P200_B256:
-        max_tokens: 200
-        variable_size: false
-        panel_size_max: 200
-        panel_size_min: 200
-        panel_selection: random
-      within_P800_B256:
-        max_tokens: 800
-        variable_size: false
-        panel_size_max: 800
-        panel_size_min: 800
-        panel_selection: random
-      within_P6400_B256:
-        max_tokens: 6400
-        variable_size: false
-        panel_size_max: 6400
-        panel_size_min: 6400
-        panel_selection: random
-      within_P_SEAAD_B256:
-        max_tokens: 800
-        variable_size: false
-        panel_size_max: 800
-        panel_size_min: 800
-        panel_selection: preselected
-        panel_filter_regex: Merfish-SEA-AD
-      within_Preselected_B256:
-        max_tokens: 800
-        variable_size: false
-        panel_size_max: 800
-        panel_size_min: 800
-        panel_selection: preselected
-        panel_filter_regex: .*
-      within_P_Xenium_hIO_B256:
-        max_tokens: 800
-        variable_size: false
-        panel_size_max: 800
-        panel_size_min: 800
-        panel_selection: preselected
-        panel_filter_regex: Xenium_hIO_v1
-      within_P_Xenium_hIO_vs_all_B256:
-        max_tokens: 1000
-        panel_overlap: true
-        variable_size: false
-        panel_size_max: 100000
-        panel_size_min: 100000
-        panel_selection: preselected
-        panel_filter_regex: Xenium_hIO_v1
-      within_P_Xenium_Human_Prime_5K_B256:
-        max_tokens: 800
-        variable_size: false
-        panel_size_max: 800
-        panel_size_min: 800
-        panel_selection: preselected
-        panel_filter_regex: Xenium_Human_Prime_5K
     train:
       max_tokens: 1000
       variable_size: false
@@ -125,79 +56,6 @@ datamodule:
       panel_max_drop_rate: 0.5
       panel_selection_mixed_prob: 0.25
   dataloader:
-    val:
-      same:
-        shuffle: true
-        drop_last: true
-        batch_size: 512
-        pin_memory: false
-        num_samples: 50000
-        num_workers: 2
-        within_group_sampling: dataset
-      within_P200_B256:
-        shuffle: true
-        drop_last: true
-        batch_size: 256
-        pin_memory: false
-        num_samples: 50000
-        num_workers: 2
-        within_group_sampling: dataset
-      within_P800_B256:
-        shuffle: true
-        drop_last: true
-        batch_size: 256
-        pin_memory: false
-        num_samples: 50000
-        num_workers: 2
-        within_group_sampling: dataset
-      within_P6400_B256:
-        shuffle: true
-        drop_last: true
-        batch_size: 256
-        pin_memory: false
-        num_samples: 50000
-        num_workers: 2
-        within_group_sampling: dataset
-      within_P_SEAAD_B256:
-        shuffle: true
-        drop_last: true
-        batch_size: 256
-        pin_memory: false
-        num_samples: 50000
-        num_workers: 2
-        within_group_sampling: dataset
-      within_Preselected_B256:
-        shuffle: true
-        drop_last: true
-        batch_size: 256
-        pin_memory: false
-        num_samples: 50000
-        num_workers: 2
-        within_group_sampling: dataset
-      within_P_Xenium_hIO_B256:
-        shuffle: true
-        drop_last: true
-        batch_size: 256
-        pin_memory: false
-        num_samples: 50000
-        num_workers: 2
-        within_group_sampling: dataset
-      within_P_Xenium_hIO_vs_all_B256:
-        shuffle: true
-        drop_last: true
-        batch_size: 256
-        pin_memory: false
-        num_samples: 50000
-        num_workers: 2
-        within_group_sampling: dataset
-      within_P_Xenium_Human_Prime_5K_B256:
-        shuffle: true
-        drop_last: true
-        batch_size: 256
-        pin_memory: false
-        num_samples: 50000
-        num_workers: 2
-        within_group_sampling: dataset
     train:
       shuffle: true
       drop_last: true