| slurm_config: big | |
| task_type: local_predict | |
| dataset: | |
| split: test | |
| video_processor: VideoProcessor | |
| aligner: MSRVTTQAAligner | |
| bert_name: bert-base-uncased | |
| meta_processor: MSRVTTQAMetaProcessor | |
| test_path: data/msrvtt-qa/MSR_MC_test.csv | |
| vfeat_dir: data/feat/feat_vtt_s3d | |
| text_processor: MSRVTTQATextProcessor | |
| num_iso_layer: 12 | |
| max_video_len: 32 | |
| max_len: 96 | |
| fairseq: | |
| dataset: | |
| batch_size: 256 | |
| valid_subset: test | |
| num_workers: 2 | |
| common_eval: | |
| path: runs/retri/videoclip/checkpoint_best.pt | |
| model: | |
| model_cls: MMFusionSeparate | |
| mm_encoder_cls: null | |
| video_encoder_cls: MMBertForEncoder | |
| text_encoder_cls: BertModel | |
| num_hidden_video_layers: 6 | |
| eval: | |
| save_path: runs/retri/videoclip/vttqa_zs/eval | |
| metric: QAMetric | |
| predictor: QAPredictor | |