| --- |
| dataset_info: |
| - config_name: backwardFacingStep2d |
| features: |
| - name: query_id |
| dtype: string |
| - name: case_id |
| dtype: string |
| - name: raw_text |
| dtype: string |
| - name: parsed_text |
| dtype: string |
| - name: video |
| dtype: video |
| splits: |
| - name: train |
| num_bytes: 11207416 |
| num_examples: 26 |
| download_size: 11207416 |
| dataset_size: 11207416 |
| - config_name: buoyancy2d |
| features: |
| - name: query_id |
| dtype: string |
| - name: case_id |
| dtype: string |
| - name: raw_text |
| dtype: string |
| - name: parsed_text |
| dtype: string |
| - name: video |
| dtype: video |
| splits: |
| - name: train |
| num_bytes: 66886996 |
| num_examples: 56 |
| download_size: 66886996 |
| dataset_size: 66886996 |
| - config_name: cavity2d_turb |
| features: |
| - name: query_id |
| dtype: string |
| - name: case_id |
| dtype: string |
| - name: raw_text |
| dtype: string |
| - name: parsed_text |
| dtype: string |
| - name: video |
| dtype: video |
| splits: |
| - name: train |
| num_bytes: 61808418 |
| num_examples: 58 |
| download_size: 61808418 |
| dataset_size: 61808418 |
| - config_name: dye2d |
| features: |
| - name: query_id |
| dtype: string |
| - name: case_id |
| dtype: string |
| - name: raw_text |
| dtype: string |
| - name: parsed_text |
| dtype: string |
| - name: video |
| dtype: video |
| splits: |
| - name: train |
| num_bytes: 7539274 |
| num_examples: 100 |
| download_size: 7539274 |
| dataset_size: 7539274 |
| - config_name: flowpastcylinder2d |
| features: |
| - name: query_id |
| dtype: string |
| - name: case_id |
| dtype: string |
| - name: raw_text |
| dtype: string |
| - name: parsed_text |
| dtype: string |
| - name: video |
| dtype: video |
| splits: |
| - name: train |
| num_bytes: 29937995 |
| num_examples: 56 |
| download_size: 29937995 |
| dataset_size: 29937995 |
| - config_name: flowpastsquare2d |
| features: |
| - name: query_id |
| dtype: string |
| - name: case_id |
| dtype: string |
| - name: raw_text |
| dtype: string |
| - name: parsed_text |
| dtype: string |
| - name: video |
| dtype: video |
| splits: |
| - name: train |
| num_bytes: 20568901 |
| num_examples: 44 |
| download_size: 20568901 |
| dataset_size: 20568901 |
| - config_name: jetaxi |
| features: |
| - name: query_id |
| dtype: string |
| - name: case_id |
| dtype: string |
| - name: raw_text |
| dtype: string |
| - name: parsed_text |
| dtype: string |
| - name: video |
| dtype: video |
| splits: |
| - name: train |
| num_bytes: 15131503 |
| num_examples: 53 |
| download_size: 15131503 |
| dataset_size: 15131503 |
| - config_name: jetaxi_turb |
| features: |
| - name: query_id |
| dtype: string |
| - name: case_id |
| dtype: string |
| - name: raw_text |
| dtype: string |
| - name: parsed_text |
| dtype: string |
| - name: video |
| dtype: video |
| splits: |
| - name: train |
| num_bytes: 13398080 |
| num_examples: 47 |
| download_size: 13398080 |
| dataset_size: 13398080 |
| - config_name: lidDrivenCavity3d |
| features: |
| - name: query_id |
| dtype: string |
| - name: case_id |
| dtype: string |
| - name: raw_text |
| dtype: string |
| - name: parsed_text |
| dtype: string |
| - name: video |
| dtype: video |
| splits: |
| - name: train |
| num_bytes: 20150317 |
| num_examples: 42 |
| download_size: 20150317 |
| dataset_size: 20150317 |
| - config_name: naca0012airfoil2d |
| features: |
| - name: query_id |
| dtype: string |
| - name: case_id |
| dtype: string |
| - name: raw_text |
| dtype: string |
| - name: parsed_text |
| dtype: string |
| - name: video |
| dtype: video |
| splits: |
| - name: train |
| num_bytes: 39326040 |
| num_examples: 100 |
| download_size: 39326040 |
| dataset_size: 39326040 |
| - config_name: rayleighBenard2d |
| features: |
| - name: query_id |
| dtype: string |
| - name: case_id |
| dtype: string |
| - name: raw_text |
| dtype: string |
| - name: parsed_text |
| dtype: string |
| - name: video |
| dtype: video |
| splits: |
| - name: train |
| num_bytes: 22114442 |
| num_examples: 44 |
| download_size: 22114442 |
| dataset_size: 22114442 |
| - config_name: suddenContraction2d |
| features: |
| - name: query_id |
| dtype: string |
| - name: case_id |
| dtype: string |
| - name: raw_text |
| dtype: string |
| - name: parsed_text |
| dtype: string |
| - name: video |
| dtype: video |
| splits: |
| - name: train |
| num_bytes: 4494372 |
| num_examples: 14 |
| download_size: 4494372 |
| dataset_size: 4494372 |
| - config_name: suddenExpansion2d |
| features: |
| - name: query_id |
| dtype: string |
| - name: case_id |
| dtype: string |
| - name: raw_text |
| dtype: string |
| - name: parsed_text |
| dtype: string |
| - name: video |
| dtype: video |
| splits: |
| - name: train |
| num_bytes: 9908677 |
| num_examples: 21 |
| download_size: 9908677 |
| dataset_size: 9908677 |
| - config_name: suddenExpansion2d_nn_hb |
| features: |
| - name: query_id |
| dtype: string |
| - name: case_id |
| dtype: string |
| - name: raw_text |
| dtype: string |
| - name: parsed_text |
| dtype: string |
| - name: video |
| dtype: video |
| splits: |
| - name: train |
| num_bytes: 7525010 |
| num_examples: 18 |
| download_size: 7525010 |
| dataset_size: 7525010 |
| - config_name: suddenExpansion2d_nn_powerlaw |
| features: |
| - name: query_id |
| dtype: string |
| - name: case_id |
| dtype: string |
| - name: raw_text |
| dtype: string |
| - name: parsed_text |
| dtype: string |
| - name: video |
| dtype: video |
| splits: |
| - name: train |
| num_bytes: 8718131 |
| num_examples: 21 |
| download_size: 8718131 |
| dataset_size: 8718131 |
| --- |
| |
| # Physics Bench Fluid Eval |
|
|
| Repository: `gowitheflowlab/physics-bench-fluid-eval` |
|
|
| Evaluation split for fluid simulation video retrieval. |
|
|
| - configs: 15, one per family, each holding `train-00000-of-00001.parquet` |
| - rows: 700, one row per benchmark case (`query_id = <case_id>__full__0`) |
| - columns: `query_id`, `case_id`, `raw_text`, `parsed_text`, embedded `video` |
| - `parsed_text` is the natural-language query and uses the same generator |
| (`fluid_eval_queries_v2_dynamics_aligned_raw_parsed`) as `gowitheflowlab/physics-bench-fluid-train`, so training and |
| evaluation queries are phrased identically |
| - `raw_text` keeps the structured single-sentence variant of the same constraints |
| - videos are H.264 and unchanged from the previous export |
| - parquet files carry one row group per case and page indexes for Dataset Viewer reads |
|
|
| Case counts per config: |
|
|
| - `backwardFacingStep2d`: 26 cases |
| - `buoyancy2d`: 56 cases |
| - `cavity2d_turb`: 58 cases |
| - `dye2d`: 100 cases |
| - `flowpastcylinder2d`: 56 cases |
| - `flowpastsquare2d`: 44 cases |
| - `jetaxi`: 53 cases |
| - `jetaxi_turb`: 47 cases |
| - `lidDrivenCavity3d`: 42 cases |
| - `naca0012airfoil2d`: 100 cases |
| - `rayleighBenard2d`: 44 cases |
| - `suddenContraction2d`: 14 cases |
| - `suddenExpansion2d`: 21 cases |
| - `suddenExpansion2d_nn_hb`: 18 cases |
| - `suddenExpansion2d_nn_powerlaw`: 21 cases |
|
|
