diff options
| author | Niclas Dobbertin <niclas.dobbertin@mailbox.org> | 2024-09-20 11:08:29 +0200 |
|---|---|---|
| committer | Niclas Dobbertin <niclas.dobbertin@mailbox.org> | 2024-09-20 11:08:29 +0200 |
| commit | 687a14059c0bd1a8b8c10cbdf82155ba57f6c31a (patch) | |
| tree | d965e764a7f1f3b41dbee2c9737fce119a25f193 /modeling/prod_numbers.py | |
| parent | bf2f74d43a02ef01c794292bf10fa5e5652447f4 (diff) | |
reward for done retrieval to prevent loop wip
Diffstat (limited to 'modeling/prod_numbers.py')
| -rw-r--r-- | modeling/prod_numbers.py | 6 |
1 files changed, 6 insertions, 0 deletions
diff --git a/modeling/prod_numbers.py b/modeling/prod_numbers.py index 3943403..773768d 100644 --- a/modeling/prod_numbers.py +++ b/modeling/prod_numbers.py @@ -28,6 +28,7 @@ def number(Model): number_expand_arg1_done = Model.productionstring( name="number_expand_arg1_done", + reward=100, string=""" =g> isa math_goal @@ -54,6 +55,7 @@ def number(Model): number_expand_arg1_fail = Model.productionstring( name="number_expand_arg1_fail", + # reward=-100, string=""" =g> isa math_goal @@ -97,6 +99,7 @@ def number(Model): number_expand_arg2_fail = Model.productionstring( name="number_expand_arg2_fail", + # reward=-100, string=""" =g> isa math_goal @@ -116,6 +119,7 @@ def number(Model): number_expand_arg2_done = Model.productionstring( name="number_expand_arg2_done", + reward=100, string=""" =g> isa math_goal @@ -142,6 +146,7 @@ def number(Model): number_expand_done = Model.productionstring( name="number_expand_done", + reward=100, string=""" =g> isa math_goal @@ -191,6 +196,7 @@ def number(Model): number_contract_result_done = Model.productionstring( name="number_contract_result_done", + reward=100, string=""" =g> isa math_goal |