Upload dataset_card.yaml

dataset_card.yaml

@@ -0,0 +1,152 @@
+dataset_name: hlo-feature-dataset
+pretty_name: HLO Feature Dataset for Deep Learning Resource Estimation
+dataset_type: graph-and-tabular
+license: MIT
+task_categories:
+- time-series-forecasting
+- regression
+- graph-machine-learning
+language: en
+tags:
+- HPC
+- resource-prediction
+- XLA
+- compiler-features
+- deep-learning
+- graph-learning
+- scheduling
+size_categories:
+- 1K<n<10K
+source_datasets:
+- custom
+dataset_summary: The HLO Feature Dataset contains High-Level Optimizer (HLO) graph
+  features and metadata extracted from deep learning training workloads. It is designed
+  for tasks such as runtime prediction, resource estimation, and graph-based machine
+  learning in HPC environments.
+structured_data:
+  features:
+  - name: name
+    type: string
+    description: ''
+  - name: samples
+    type: float
+    description: ''
+  - name: input_dim_w
+    type: float
+    description: ''
+  - name: input_dim_h
+    type: float
+    description: ''
+  - name: input_dim_c
+    type: float
+    description: ''
+  - name: output_dim
+    type: float
+    description: ''
+  - name: optimizer
+    type: string
+    description: ''
+  - name: epochs
+    type: float
+    description: ''
+  - name: batch
+    type: float
+    description: ''
+  - name: learn_rate
+    type: float
+    description: ''
+  - name: tf_version
+    type: string
+    description: ''
+  - name: cuda_version
+    type: string
+    description: ''
+  - name: batch_time
+    type: float
+    description: ''
+  - name: epoch_time
+    type: float
+    description: ''
+  - name: fit_time
+    type: float
+    description: ''
+  - name: npz_path
+    type: string
+    description: ''
+  - name: gpu_make
+    type: string
+    description: ''
+  - name: gpu_name
+    type: string
+    description: ''
+  - name: gpu_arch
+    type: string
+    description: ''
+  - name: gpu_cc
+    type: string
+    description: ''
+  - name: gpu_core_count
+    type: string
+    description: ''
+  - name: gpu_sm_count
+    type: string
+    description: ''
+  - name: gpu_memory_size
+    type: string
+    description: ''
+  - name: gpu_memory_type
+    type: string
+    description: ''
+  - name: gpu_memory_bw
+    type: string
+    description: ''
+  - name: gpu_tensor_core_count
+    type: string
+    description: ''
+  - name: max_memory_util
+    type: float
+    description: ''
+  - name: avg_memory_util
+    type: float
+    description: ''
+  - name: max_gpu_util
+    type: string
+    description: ''
+  - name: avg_gpu_util
+    type: string
+    description: ''
+  - name: max_gpu_temp
+    type: string
+    description: ''
+  - name: avg_gpu_temp
+    type: string
+    description: ''
+graph_data:
+  node_features: node_feat
+  edge_index: edge_index
+  additional_keys:
+  - node_opcode
+  - node_config_ids
+  - node_splits
+usage_example: '```python
+
+  from datasets import load_dataset
+
+  import numpy as np
+
+
+  dataset = load_dataset(''your-username/hlo-feature-dataset'')
+
+  sample = dataset[''train''][0]
+
+
+  graph_data = np.load(sample[''npz_path''])
+
+  node_features = graph_data[''node_feat'']
+
+  edges = graph_data[''edge_index'']
+
+  ```'
+citation: "@misc{hlofeatures2025,\n  title={HLO Feature Dataset for AI Resource Estimation},\n\
+  \  author={Your Name},\n  year={2025},\n  url={https://huggingface.co/datasets/your-username/hlo-feature-dataset}\n\
+  }"