Merge branch 'main' of https://github.com/zou-group/textgrad

README.md

@@ -54,7 +54,7 @@ TextGrad can optimize unstructured variables, such as text. Let us have an initi
 
 ```python
 import textgrad as tg
-tg.set_backward_engine(tg.get_engine("gpt-4o"))
+tg.set_backward_engine("gpt-4o")
 
 initial_solution = """To solve the equation 3x^2 - 7x + 2 = 0, we use the quadratic formula:
 x = (-b ± √(b^2 - 4ac)) / 2a
@@ -116,7 +116,7 @@ TextGrad can also optimize prompts in PyTorch style! Here's how to do it with Te
 ```python
 import textgrad as tg
 llm_engine = tg.get_engine("gpt-3.5-turbo")
-tg.set_backward_engine(tg.get_engine("gpt-4o"))
+tg.set_backward_engine("gpt-4o")
 
 _, val_set, _, eval_fn = load_task("BBH_object_counting", llm_engine)
 question_str, answer_str = val_set[0]

evaluation/prompt_optimization.py

@@ -16,7 +16,7 @@ def set_seed(seed):
 
 def config():
     parser = argparse.ArgumentParser(description="Optimize a prompt for a task.")
-    parser.add_argument("--task", type=str, default=" ", help="The task to evaluate the model on.")
+    parser.add_argument("--task", type=str, default="BBH_object_counting", help="The task to evaluate the model on.")
     parser.add_argument("--evaluation_engine", type=str, default="gpt-4o", help="The API to use for evaluation.")
     parser.add_argument("--test_engine", type=str, default="gpt-3.5-turbo-0125", help="The API to use for evaluation.")
     parser.add_argument("--batch_size", type=int, default=3, help="The batch size to use for training.")