Add seq2seq eval benchmark callback (#1274)
* Add CausalLMBenchEvalCallback for measuring seq2seq performance * Fix code for pre-commit * Fix typing and improve logging * eval_sample_packing must be false with CausalLMBenchEvalCallback
This commit is contained in:
@@ -23,7 +23,7 @@ numba
|
||||
numpy>=1.24.4
|
||||
mlflow
|
||||
# qlora things
|
||||
evaluate==0.4.0
|
||||
evaluate==0.4.1
|
||||
scipy
|
||||
scikit-learn==1.2.2
|
||||
pynvml
|
||||
|
||||
Reference in New Issue
Block a user