DeepTalk-singleST.ipynb tutorial not running

[kmccrock]

Hello, I get an error trying to convert tensor to numpy from dt.Train. See below.

I installed my environment according to the GitHub readme and just confirmed my versions match. I have not been able to install from the environment.yml file though, I get version compatibility errors.

I modified your code to handle this conversion, but now I do not get the .model file which I need for dt.run_predict.

Thanks!
-Kyle

File ~/Desktop/DeepTalk Tutorial/DeepTalk-main/DeepTalk_ST/src/utils/evaluation.py:105, in run_evaluation_main(test_edges, prediction_data, true_data, threshold, header)
102 y_pred[i] = 1
104 y_true = np.array(true_data)
--> 105 y_scores = np.array(prediction_data)
107 ps, rs, _ = precision_recall_curve(y_true, y_scores)
...
--> 621 return self.numpy()
622 else:
623 return self.numpy().astype(dtype, copy=False)

TypeError: can't convert cuda:0 device type tensor to numpy. Use Tensor.cpu() to copy the tensor to host memory first.