about openai apiky usage

[demaxiya567]

how can i use the“from langchain_openai import ChatOpenAI” by connect GPT4 and chemlift？when I run the example code by my API key，it always shows this error“ValueError: Sample larger than population or is negative”

ValueError Traceback (most recent call last)
Cell In[6], line 24
21 fsc = FewShotClassifier(client, property_name='bandgap')
23 # Train on a few examples
---> 24 fsc.fit(['ethane', 'propane', 'butane'], [0,1,0])
26 # Predict on a few more
27 fsc.predict(['pentane', 'hexane', 'heptane'])

File c:\Users\19341.conda\envs\agents20240401\Lib\site-packages\chemlift\icl\fewshotpredictor.py:238, in FewShotPredictor.fit(self, X, y)
231 def fit(self, X: ArrayLike, y: ArrayLike):
232 """Fit the model to a support set.
233
234 Args:
235 X: A list of examples.
236 y: A list of targets.
237 """
--> 238 support_indices = self._pick_support_indices(X)
239 self._support_set = (np.array(X)[support_indices], np.array(y)[support_indices])
240 self._allowed_values = set(y)

File c:\Users\19341.conda\envs\agents20240401\Lib\site-packages\chemlift\icl\fewshotpredictor.py:118, in FewShotPredictor._pick_support_indices(self, examples)
116 random.seed(self._seed)
117 if self._strategy == Strategy.RANDOM:
--> 118 return random.sample(range(len(examples)), self._n_support)
119 elif self._strategy == Strategy.FIRST:
120 return list(range(self._n_support))

File c:\Users\19341.conda\envs\agents20240401\Lib\random.py:456, in Random.sample(self, population, k, counts)
454 randbelow = self._randbelow
455 if not 0 <= k <= n:
--> 456 raise ValueError("Sample larger than population or is negative")
457 result = [None] * k
458 setsize = 21 # size of a small set minus size of an empty list

ValueError: Sample larger than population or is negative