Type Error #8
Description
I run python -m examples.maas.optimize --dataset GSM8K --round 1 --sample 4 --exec_model_name "gpt-4o-mini"
But it seems that some variables were of the wrong type. Do you have any idea how to fix it?
My error log:
2025-07-01 18:32:31.525 | INFO | maas.ext.maas.scripts.optimizer:_optimize_graph_maas:110 - maas/ext/maas/scripts/optimized/GSM8K/train/round_1
2025-07-01 18:32:32.137 | INFO | maas.ext.maas.benchmark.benchmark:evaluate_all_problems:124 - Starting training repetition 1/4
Repetition 1: Executing batch 1: 0%| | 0/4 [00:00<?, ?it/s]huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Repetition 1: Executing batch 1: 25%|███████████████████████████████████▎ | 1/4 [00:12<00:36, 12.08s/it]huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Repetition 1: Executing batch 1: 50%|██████████████████████████████████████████████████████████████████████▌ | 2/4 [00:12<00:10, 5.35s/it]huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Repetition 1: Executing batch 1: 100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 4/4 [00:47<00:00, 11.99s/it]
2025-07-01 18:33:20.115 | INFO | maas.ext.maas.scripts.optimizer:optimize:84 - Error occurred: expected Tensor as element 0 in argument 0, but got float. Retrying... (Attempt 1/1)
2025-07-01 18:33:20.116 | INFO | maas.ext.maas.scripts.optimizer:optimize:86 - Max retries reached. Moving to next round.
2025-07-01 18:33:25.121 | INFO | maas.ext.maas.scripts.optimizer:optimize:96 - Score for round 1: None