Datasets:

---

happynew111

/

ADARFT-wandb

like 0

Error code:   FeaturesError
Exception:    ArrowInvalid
Message:      Schema at index 1 was different: 
os: string
python: string
startedAt: string
args: list<item: string>
program: string
root: string
host: string
executable: string
cpu_count: int64
cpu_count_logical: int64
gpu: string
gpu_count: int64
disk: struct</: struct<total: string, used: string>>
memory: struct<total: string>
cpu: struct<count: int64, countLogical: int64>
gpu_nvidia: list<item: struct<name: string, memoryTotal: string, cudaCores: int64, architecture: string, uuid: string>>
cudaVersion: string
vs
_timestamp: double
_runtime: double
val-core/math_500/reward/best@1/mean: double
val-aux/aime24/reward/mean@8: double
val-aux/aime24/reward/worst@2/std: double
val-aux/aime24/reward/worst@4/std: double
_step: int64
val-aux/openai/gsm8k/reward/mean@1: double
val-aux/openai/gsm8k/reward/std@1: int64
val-aux/openai/gsm8k/reward/worst@1/mean: double
val-aux/amc23/reward/worst@2/std: double
val-aux/aime24/reward/std@8: double
val-aux/olympiadbench/reward/mean@1: double
val-core/olympiadbench/reward/best@1/mean: double
val-core/math_500/reward/best@1/std: int64
val-aux/amc23/reward/worst@4/mean: double
val-aux/amc23/reward/worst@4/std: double
val-aux/aime24/reward/best@4/std: double
val-core/aime24/reward/best@8/mean: double
val-aux/minerva_math/reward/mean@1: double
val-core/minerva_math/reward/best@1/mean: double
val-aux/math_500/reward/mean@1: double
val-aux/math_500/reward/std@1: int64
val-aux/amc23/reward/std@8: double
val-aux/amc23/reward/best@2/mean: double
val-aux/aime24/reward/best@4/mean: double
val-core/openai/gsm8k/reward/best@1/std: int64
val-aux/openai/gsm8k/reward/worst@1/std: int64
val-aux/math_500/reward/worst@1/mean: double
val-aux/amc23/reward/worst@8/mean: double
val-aux/aime24/reward/worst@4/mean: double
val-aux/aime24/reward/worst@8/std: double
val-aux/minerva_math/reward/std@1: int64
val-aux/minerva_math/reward/worst@1/mean: double
val-aux/amc23/reward/best@2/std: double
val-aux/amc23/reward/worst@2/mean: double
val-aux/amc23/reward/best@4/std: double
val-core/amc23/reward/best@8/std: double
val-aux/amc23/reward/worst@8/std: double
val-aux/aime24/reward/worst@2/mean: double
val-aux/minerva_math/reward/worst@1/std: int64
val-aux/olympiadbench/reward/std@1: int64
val-core/openai/gsm8k/reward/best@1/mean: double
val-aux/math_500/reward/worst@1/std: int64
val-aux/amc23/reward/mean@8: double
val-aux/amc23/reward/best@4/mean: double
val-core/amc23/reward/best@8/mean: double
val-aux/aime24/reward/best@2/mean: double
val-aux/aime24/reward/worst@8/mean: double
val-aux/olympiadbench/reward/worst@1/mean: double
val-aux/aime24/reward/best@2/std: double
val-core/aime24/reward/best@8/std: double
val-core/minerva_math/reward/best@1/std: int64
val-core/olympiadbench/reward/best@1/std: int64
val-aux/olympiadbench/reward/worst@1/std: int64
timing_s/adv: double
timing_s/update_actor: double
timing_s/step: double
timing_per_token_ms/update_actor: double
perf/total_num_tokens: int64
perf/time_per_step: double
global_seqlen/max: int64
global_seqlen/balanced_max: int64
actor/lr: double
critic/score/min: double
critic/rewards/mean: double
timing_s/gen: double
timing_s/old_log_prob: double
timing_per_token_ms/ref: double
actor/pg_clipfrac_lower: int64
critic/returns/min: double
global_seqlen/min: int64
global_seqlen/minmax_diff: int64
actor/kl_loss: double
actor/grad_norm: double
response_length/clip_ratio: double
prompt_length/min: int64
actor/pg_clipfrac: double
actor/ppo_kl: double
perf/mfu/actor: double
perf/cpu_memory_used_gb: double
global_seqlen/balanced_min: int64
actor/kl_coef: double
critic/score/mean: double
critic/score/max: double
critic/advantages/mean: double
critic/returns/max: double
prompt_length/mean: double
prompt_length/max: int64
global_seqlen/mean: double
actor/entropy: double
perf/max_memory_allocated_gb: double
perf/max_memory_reserved_gb: double
critic/advantages/min: double
prompt_length/clip_ratio: int64
timing_s/ref: double
timing_per_token_ms/adv: double
critic/rewards/min: double
critic/advantages/max: double
critic/returns/mean: double
response_length/mean: double
response_length/max: int64
response_length/min: int64
timing_per_token_ms/gen: double
perf/throughput: double
actor/pg_loss: double
critic/rewards/max: double
timing_s/testing: double
timing_s/save_checkpoint: double
Traceback:    Traceback (most recent call last):
                File "/src/services/worker/src/worker/job_runners/split/first_rows.py", line 228, in compute_first_rows_from_streaming_response
                  iterable_dataset = iterable_dataset._resolve_features()
                                     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/src/services/worker/.venv/lib/python3.12/site-packages/datasets/iterable_dataset.py", line 3496, in _resolve_features
                  features = _infer_features_from_batch(self.with_format(None)._head())
                                                        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/src/services/worker/.venv/lib/python3.12/site-packages/datasets/iterable_dataset.py", line 2257, in _head
                  return next(iter(self.iter(batch_size=n)))
                         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/src/services/worker/.venv/lib/python3.12/site-packages/datasets/iterable_dataset.py", line 2461, in iter
                  for key, example in iterator:
                                      ^^^^^^^^
                File "/src/services/worker/.venv/lib/python3.12/site-packages/datasets/iterable_dataset.py", line 1952, in __iter__
                  for key, pa_table in self._iter_arrow():
                                       ^^^^^^^^^^^^^^^^^^
                File "/src/services/worker/.venv/lib/python3.12/site-packages/datasets/iterable_dataset.py", line 1974, in _iter_arrow
                  yield from self.ex_iterable._iter_arrow()
                File "/src/services/worker/.venv/lib/python3.12/site-packages/datasets/iterable_dataset.py", line 531, in _iter_arrow
                  yield new_key, pa.Table.from_batches(chunks_buffer)
                                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "pyarrow/table.pxi", line 5039, in pyarrow.lib.Table.from_batches
                File "pyarrow/error.pxi", line 155, in pyarrow.lib.pyarrow_internal_check_status
                File "pyarrow/error.pxi", line 92, in pyarrow.lib.check_status
              pyarrow.lib.ArrowInvalid: Schema at index 1 was different: 
              os: string
              python: string
              startedAt: string
              args: list<item: string>
              program: string
              root: string
              host: string
              executable: string
              cpu_count: int64
              cpu_count_logical: int64
              gpu: string
              gpu_count: int64
              disk: struct</: struct<total: string, used: string>>
              memory: struct<total: string>
              cpu: struct<count: int64, countLogical: int64>
              gpu_nvidia: list<item: struct<name: string, memoryTotal: string, cudaCores: int64, architecture: string, uuid: string>>
              cudaVersion: string
              vs
              _timestamp: double
              _runtime: double
              val-core/math_500/reward/best@1/mean: double
              val-aux/aime24/reward/mean@8: double
              val-aux/aime24/reward/worst@2/std: double
              val-aux/aime24/reward/worst@4/std: double
              _step: int64
              val-aux/openai/gsm8k/reward/mean@1: double
              val-aux/openai/gsm8k/reward/std@1: int64
              val-aux/openai/gsm8k/reward/worst@1/mean: double
              val-aux/amc23/reward/worst@2/std: double
              val-aux/aime24/reward/std@8: double
              val-aux/olympiadbench/reward/mean@1: double
              val-core/olympiadbench/reward/best@1/mean: double
              val-core/math_500/reward/best@1/std: int64
              val-aux/amc23/reward/worst@4/mean: double
              val-aux/amc23/reward/worst@4/std: double
              val-aux/aime24/reward/best@4/std: double
              val-core/aime24/reward/best@8/mean: double
              val-aux/minerva_math/reward/mean@1: double
              val-core/minerva_math/reward/best@1/mean: double
              val-aux/math_500/reward/mean@1: double
              val-aux/math_500/reward/std@1: int64
              val-aux/amc23/reward/std@8: double
              val-aux/amc23/reward/best@2/mean: double
              val-aux/aime24/reward/best@4/mean: double
              val-core/openai/gsm8k/reward/best@1/std: int64
              val-aux/openai/gsm8k/reward/worst@1/std: int64
              val-aux/math_500/reward/worst@1/mean: double
              val-aux/amc23/reward/worst@8/mean: double
              val-aux/aime24/reward/worst@4/mean: double
              val-aux/aime24/reward/worst@8/std: double
              val-aux/minerva_math/reward/std@1: int64
              val-aux/minerva_math/reward/worst@1/mean: double
              val-aux/amc23/reward/best@2/std: double
              val-aux/amc23/reward/worst@2/mean: double
              val-aux/amc23/reward/best@4/std: double
              val-core/amc23/reward/best@8/std: double
              val-aux/amc23/reward/worst@8/std: double
              val-aux/aime24/reward/worst@2/mean: double
              val-aux/minerva_math/reward/worst@1/std: int64
              val-aux/olympiadbench/reward/std@1: int64
              val-core/openai/gsm8k/reward/best@1/mean: double
              val-aux/math_500/reward/worst@1/std: int64
              val-aux/amc23/reward/mean@8: double
              val-aux/amc23/reward/best@4/mean: double
              val-core/amc23/reward/best@8/mean: double
              val-aux/aime24/reward/best@2/mean: double
              val-aux/aime24/reward/worst@8/mean: double
              val-aux/olympiadbench/reward/worst@1/mean: double
              val-aux/aime24/reward/best@2/std: double
              val-core/aime24/reward/best@8/std: double
              val-core/minerva_math/reward/best@1/std: int64
              val-core/olympiadbench/reward/best@1/std: int64
              val-aux/olympiadbench/reward/worst@1/std: int64
              timing_s/adv: double
              timing_s/update_actor: double
              timing_s/step: double
              timing_per_token_ms/update_actor: double
              perf/total_num_tokens: int64
              perf/time_per_step: double
              global_seqlen/max: int64
              global_seqlen/balanced_max: int64
              actor/lr: double
              critic/score/min: double
              critic/rewards/mean: double
              timing_s/gen: double
              timing_s/old_log_prob: double
              timing_per_token_ms/ref: double
              actor/pg_clipfrac_lower: int64
              critic/returns/min: double
              global_seqlen/min: int64
              global_seqlen/minmax_diff: int64
              actor/kl_loss: double
              actor/grad_norm: double
              response_length/clip_ratio: double
              prompt_length/min: int64
              actor/pg_clipfrac: double
              actor/ppo_kl: double
              perf/mfu/actor: double
              perf/cpu_memory_used_gb: double
              global_seqlen/balanced_min: int64
              actor/kl_coef: double
              critic/score/mean: double
              critic/score/max: double
              critic/advantages/mean: double
              critic/returns/max: double
              prompt_length/mean: double
              prompt_length/max: int64
              global_seqlen/mean: double
              actor/entropy: double
              perf/max_memory_allocated_gb: double
              perf/max_memory_reserved_gb: double
              critic/advantages/min: double
              prompt_length/clip_ratio: int64
              timing_s/ref: double
              timing_per_token_ms/adv: double
              critic/rewards/min: double
              critic/advantages/max: double
              critic/returns/mean: double
              response_length/mean: double
              response_length/max: int64
              response_length/min: int64
              timing_per_token_ms/gen: double
              perf/throughput: double
              actor/pg_loss: double
              critic/rewards/max: double
              timing_s/testing: double
              timing_s/save_checkpoint: double