Datasets:

optimum-benchmark

/

llm-perf-leaderboard

like 0

Error code:   DatasetGenerationCastError
Exception:    DatasetGenerationCastError
Message:      An error occurred while generating the dataset

All the data files must have the same columns, but at some point there are 150 new columns ({'config.backend.model', 'report.prefill.efficiency.value', 'report.prefill.latency.p50', 'config.backend.torch_compile_target', 'config.launcher.name', 'report.decode.efficiency.value', 'config.environment.timm_version', 'config.scenario._target_', 'config.environment.cpu_count', 'config.environment.gpu_vram_mb', 'report.prefill.memory.max_global_vram', 'report.per_token.latency.p50', 'report.decode.latency.stdev', 'config.backend.hub_kwargs.force_download', 'config.scenario.name', 'report.prefill.latency.stdev', 'report.prefill.memory.max_process_vram', 'report.per_token.throughput.value', 'config.environment.machine', 'config.launcher._target_', 'config.environment.platform', 'report.prefill.latency.total', 'config.backend.autocast_enabled', 'report.decode.latency.p95', 'report.per_token.efficiency', 'report.per_token.memory', 'config.environment.transformers_commit', 'report.prefill.throughput.unit', 'config.environment.diffusers_commit', 'config.scenario.latency', 'config.environment.accelerate_commit', 'report.per_token.latency.count', 'config.backend.torch_dtype', 'config.backend.to_bettertransformer', 'config.launcher.start_method', 'config.environment.optimum_benchmark_commit', 'config.backend.cache_implementation', 'report.prefill.latency.mean', 'config.backend.hub_kwargs.revision', 'report.decode.latency.unit', 'config.environment.diffusers_version', 'report.prefill.memory.max_reserved', 'report.per_token.latency.p99', 'config.environment.transformers_version', 're
...
nt.cpu', 'config.environment.peft_version', 'config.environment.peft_commit', 'config.launcher.device_isolation_action', 'config.scenario.generate_kwargs.min_new_tokens', 'config.environment.optimum_benchmark_version', 'config.launcher.numactl', 'config.environment.gpu', 'config.backend.quantization_scheme', 'config.environment.python_version', 'report.prefill.energy.ram', 'report.decode.energy.cpu', 'config.scenario.input_shapes.batch_size', 'config.backend.library', 'report.prefill.throughput.value', 'report.prefill.efficiency.unit', 'report.decode.latency.p99', 'report.decode.latency.values', 'report.prefill.energy.unit', 'report.decode.memory.max_ram', 'config.environment.timm_commit', 'config.backend.quantization_config.version', 'report.decode.latency.p90', 'config.backend.quantization_config.bits', 'config.environment.system', 'report.decode.efficiency.unit', 'config.scenario.energy', 'config.backend.hub_kwargs.trust_remote_code', 'report.per_token.latency.total', 'config.launcher.device_isolation', 'report.decode.memory.max_allocated', 'report.prefill.latency.values', 'config.backend.inter_op_num_threads', 'config.environment.optimum_commit', 'config.scenario.input_shapes.sequence_length', 'config.environment.gpu_count', 'report.decode.memory.max_global_vram', 'config.backend.processor', 'config.backend.torch_compile', 'config.backend.no_weights', 'report.per_token.latency.stdev', 'report.prefill.latency.p95', 'report.prefill.latency.p99', 'report.prefill.energy.cpu'}) and 21 missing columns ({'Model', 'MoE', 'Architecture', 'GSM8K', 'HellaSwag', 'Precision', 'T', 'Hub License', 'TruthfulQA', 'Flagged', 'Model sha', 'Weight type', 'MMLU', 'Merged', 'Average ⬆️', 'Winogrande', 'Available on the hub', 'Type', 'Hub ❤️', 'ARC', '#Params (B)'}).

This happened while the csv dataset builder was generating data using

hf://datasets/optimum-benchmark/llm-perf-leaderboard/perf-df-awq-1xA10.csv (at revision eb9b1fab2d755287ab732703dfbb1e4b47719cc2)

Please either edit the data files to have matching columns, or separate them into different configurations (see docs at https://hf.co/docs/hub/datasets-manual-configuration#multiple-configurations)
Traceback:    Traceback (most recent call last):
                File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/builder.py", line 2011, in _prepare_split_single
                  writer.write_table(table)
                File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/arrow_writer.py", line 585, in write_table
                  pa_table = table_cast(pa_table, self._schema)
                File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/table.py", line 2302, in table_cast
                  return cast_table_to_schema(table, schema)
                File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/table.py", line 2256, in cast_table_to_schema
                  raise CastError(
              datasets.table.CastError: Couldn't cast
              config.name: string
              config.backend.name: string
              config.backend.version: string
              config.backend._target_: string
              config.backend.task: string
              config.backend.library: string
              config.backend.model: string
              config.backend.processor: string
              config.backend.device: string
              config.backend.device_ids: int64
              config.backend.seed: int64
              config.backend.inter_op_num_threads: double
              config.backend.intra_op_num_threads: double
              config.backend.model_kwargs.trust_remote_code: bool
              config.backend.processor_kwargs.trust_remote_code: bool
              config.backend.hub_kwargs.trust_remote_code: bool
              config.backend.no_weights: bool
              config.backend.device_map: double
              config.backend.torch_dtype: string
              config.backend.eval_mode: bool
              config.backend.to_bettertransformer: bool
              config.backend.low_cpu_mem_usage: double
              config.backend.attn_implementation: string
              config.backend.cache_implementation: double
              config.backend.autocast_enabled: bool
              config.backend.autocast_dtype: double
              config.backend.torch_compile: bool
              config.backend.torch_compile_target: string
              config.backend.quantization_scheme: string
              config.backend.quantization_config.bits: int64
              config.backend.quantization_config.version: string
              config.backend.deepspeed_inference: bool
              config.backend.peft_type: double
              config.scenario.name: string
              config.scenario._target_: string
              config.scenario.iterations: int64
              config.scenario.duration: int64
              config.scenario.warmup_runs: int64
              config.scenario.input_shapes.batch_size: int64
              config.scenario.input_shapes.num_choices: int64
              co
              ...
              .latency.p50: double
              report.decode.latency.p90: double
              report.decode.latency.p95: double
              report.decode.latency.p99: double
              report.decode.latency.values: string
              report.decode.throughput.unit: string
              report.decode.throughput.value: double
              report.decode.energy.unit: string
              report.decode.energy.cpu: double
              report.decode.energy.ram: double
              report.decode.energy.gpu: double
              report.decode.energy.total: double
              report.decode.efficiency.unit: string
              report.decode.efficiency.value: double
              report.per_token.memory: double
              report.per_token.latency.unit: string
              report.per_token.latency.count: double
              report.per_token.latency.total: double
              report.per_token.latency.mean: double
              report.per_token.latency.stdev: double
              report.per_token.latency.p50: double
              report.per_token.latency.p90: double
              report.per_token.latency.p95: double
              report.per_token.latency.p99: double
              report.per_token.latency.values: string
              report.per_token.throughput.unit: string
              report.per_token.throughput.value: double
              report.per_token.energy: double
              report.per_token.efficiency: double
              config.backend.hub_kwargs.revision: string
              config.backend.hub_kwargs.force_download: bool
              config.backend.hub_kwargs.local_files_only: bool
              config.backend.quantization_config.exllama_config.version: double
              config.backend.quantization_config.exllama_config.max_input_len: double
              config.backend.quantization_config.exllama_config.max_batch_size: double
              -- schema metadata --
              pandas: '{"index_columns": [{"kind": "range", "name": null, "start": 0, "' + 24248
              to
              {'T': Value(dtype='string', id=None), 'Model': Value(dtype='string', id=None), 'Average ⬆️': Value(dtype='float64', id=None), 'ARC': Value(dtype='float64', id=None), 'HellaSwag': Value(dtype='float64', id=None), 'MMLU': Value(dtype='float64', id=None), 'TruthfulQA': Value(dtype='float64', id=None), 'Winogrande': Value(dtype='float64', id=None), 'GSM8K': Value(dtype='float64', id=None), 'Type': Value(dtype='string', id=None), 'Architecture': Value(dtype='string', id=None), 'Weight type': Value(dtype='string', id=None), 'Precision': Value(dtype='string', id=None), 'Merged': Value(dtype='bool', id=None), 'Hub License': Value(dtype='string', id=None), '#Params (B)': Value(dtype='float64', id=None), 'Hub ❤️': Value(dtype='float64', id=None), 'Available on the hub': Value(dtype='bool', id=None), 'Model sha': Value(dtype='string', id=None), 'Flagged': Value(dtype='bool', id=None), 'MoE': Value(dtype='bool', id=None)}
              because column names don't match
              
              During handling of the above exception, another exception occurred:
              
              Traceback (most recent call last):
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 1324, in compute_config_parquet_and_info_response
                  parquet_operations = convert_to_parquet(builder)
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 938, in convert_to_parquet
                  builder.download_and_prepare(
                File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/builder.py", line 1027, in download_and_prepare
                  self._download_and_prepare(
                File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/builder.py", line 1122, in _download_and_prepare
                  self._prepare_split(split_generator, **prepare_split_kwargs)
                File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/builder.py", line 1882, in _prepare_split
                  for job_id, done, content in self._prepare_split_single(
                File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/builder.py", line 2013, in _prepare_split_single
                  raise DatasetGenerationCastError.from_cast_error(
              datasets.exceptions.DatasetGenerationCastError: An error occurred while generating the dataset
              
              All the data files must have the same columns, but at some point there are 150 new columns ({'config.backend.model', 'report.prefill.efficiency.value', 'report.prefill.latency.p50', 'config.backend.torch_compile_target', 'config.launcher.name', 'report.decode.efficiency.value', 'config.environment.timm_version', 'config.scenario._target_', 'config.environment.cpu_count', 'config.environment.gpu_vram_mb', 'report.prefill.memory.max_global_vram', 'report.per_token.latency.p50', 'report.decode.latency.stdev', 'config.backend.hub_kwargs.force_download', 'config.scenario.name', 'report.prefill.latency.stdev', 'report.prefill.memory.max_process_vram', 'report.per_token.throughput.value', 'config.environment.machine', 'config.launcher._target_', 'config.environment.platform', 'report.prefill.latency.total', 'config.backend.autocast_enabled', 'report.decode.latency.p95', 'report.per_token.efficiency', 'report.per_token.memory', 'config.environment.transformers_commit', 'report.prefill.throughput.unit', 'config.environment.diffusers_commit', 'config.scenario.latency', 'config.environment.accelerate_commit', 'report.per_token.latency.count', 'config.backend.torch_dtype', 'config.backend.to_bettertransformer', 'config.launcher.start_method', 'config.environment.optimum_benchmark_commit', 'config.backend.cache_implementation', 'report.prefill.latency.mean', 'config.backend.hub_kwargs.revision', 'report.decode.latency.unit', 'config.environment.diffusers_version', 'report.prefill.memory.max_reserved', 'report.per_token.latency.p99', 'config.environment.transformers_version', 're
              ...
              nt.cpu', 'config.environment.peft_version', 'config.environment.peft_commit', 'config.launcher.device_isolation_action', 'config.scenario.generate_kwargs.min_new_tokens', 'config.environment.optimum_benchmark_version', 'config.launcher.numactl', 'config.environment.gpu', 'config.backend.quantization_scheme', 'config.environment.python_version', 'report.prefill.energy.ram', 'report.decode.energy.cpu', 'config.scenario.input_shapes.batch_size', 'config.backend.library', 'report.prefill.throughput.value', 'report.prefill.efficiency.unit', 'report.decode.latency.p99', 'report.decode.latency.values', 'report.prefill.energy.unit', 'report.decode.memory.max_ram', 'config.environment.timm_commit', 'config.backend.quantization_config.version', 'report.decode.latency.p90', 'config.backend.quantization_config.bits', 'config.environment.system', 'report.decode.efficiency.unit', 'config.scenario.energy', 'config.backend.hub_kwargs.trust_remote_code', 'report.per_token.latency.total', 'config.launcher.device_isolation', 'report.decode.memory.max_allocated', 'report.prefill.latency.values', 'config.backend.inter_op_num_threads', 'config.environment.optimum_commit', 'config.scenario.input_shapes.sequence_length', 'config.environment.gpu_count', 'report.decode.memory.max_global_vram', 'config.backend.processor', 'config.backend.torch_compile', 'config.backend.no_weights', 'report.per_token.latency.stdev', 'report.prefill.latency.p95', 'report.prefill.latency.p99', 'report.prefill.energy.cpu'}) and 21 missing columns ({'Model', 'MoE', 'Architecture', 'GSM8K', 'HellaSwag', 'Precision', 'T', 'Hub License', 'TruthfulQA', 'Flagged', 'Model sha', 'Weight type', 'MMLU', 'Merged', 'Average ⬆️', 'Winogrande', 'Available on the hub', 'Type', 'Hub ❤️', 'ARC', '#Params (B)'}).
              
              This happened while the csv dataset builder was generating data using
              
              hf://datasets/optimum-benchmark/llm-perf-leaderboard/perf-df-awq-1xA10.csv (at revision eb9b1fab2d755287ab732703dfbb1e4b47719cc2)
              
              Please either edit the data files to have matching columns, or separate them into different configurations (see docs at https://hf.co/docs/hub/datasets-manual-configuration#multiple-configurations)