QNNs for Stock Price Trend Prediction
Develop quantum neural network models using PyTorch to predict stock price trends based on limit order book data.
Hosted by
Quantum Signals
Posted
Getting an error when running the dummy model
Hi there, I am getting below error when running the notebook.
---------------------------------------------------------------------------
TypeError Traceback (most recent call last)
Cell In[4], line 2
1 # Get predictions from the model
----> 2 predictions = model(torch.Tensor(input[0]))
4 # Move predictions to CPU and convert to NumPy array
5 predictions = predictions.cpu().numpy()
TypeError: 'method' object is not subscriptableI believe maybe it is because the notebook uses
input as a variable?# Get predictions from the model
predictions = model(torch.Tensor(input[0]))
# Move predictions to CPU and convert to NumPy array
predictions = predictions.cpu().numpy()Anyways I tried to change that part and run
aqora test but got another error.# Example input to test the model
sample_input = torch.rand((1, 10)) # Random input with 10 features
# Get predictions from the model
predictions = model(sample_input)
# Move predictions to CPU and convert to NumPy array
predictions = predictions.cpu().detach().numpy()Order by:
Posted by antoine •
Hello @hasarinduperera
Hi there, I am getting below error when running the notebook.
Are you running the notebook with
aqora test as expected? Or do you run it using Visual Studio Code?Anyways I tried to change that part and runaqora testbut got another error.
Can you quote the full error trace in the topic please?
Posted by hasarinduperera •
Are you running the notebook with aqora test as expected? Or do you run it using Visual Studio Code?
aqora test worked fine on terminal. But then I tried to follow the notebook (running cell by cell). That's when I got the error.Can you quote the full error trace in the topic please?
Model output: [[1 1 2 1 2]]
Model output: [[1 1 2 1 2]]
[1 FAIL] Evaluation error
Model output: [[1 1 2 1 2]]
Traceback (most recent call last):
File "/home/personal/Documents/Personal/Documents/aqora/quantum-signals-lob/.venv/lib/python3.10/site-packages/use_case/run.py", line 32, in score
return compute_score(Tensor(solution), Tensor(original_input[1]))
File "/home/personal/Documents/Personal/Documents/aqora/quantum-signals-lob/.venv/lib/python3.10/site-packages/use_case/metrics.py", line 54, in compute_score
scores = [
File "/home/personal/Documents/Personal/Documents/aqora/quantum-signals-lob/.venv/lib/python3.10/site-packages/use_case/metrics.py", line 55, in <listcomp>
compute_metrics(truth[:, h], prediction[:, h])[cst.Metrics.F1.value]
File "/home/personal/Documents/Personal/Documents/aqora/quantum-signals-lob/.venv/lib/python3.10/site-packages/use_case/metrics.py", line 19, in compute_metrics
cr = classification_report(truth, prediction, output_dict=True, zero_division=0)
File "/home/personal/Documents/Personal/Documents/aqora/quantum-signals-lob/.venv/lib/python3.10/site-packages/sklearn/utils/_param_validation.py", line 213, in wrapper
return func(*args, **kwargs)
File "/home/personal/Documents/Personal/Documents/aqora/quantum-signals-lob/.venv/lib/python3.10/site-packages/sklearn/metrics/_classification.py", line 2626, in classification_report
y_type, y_true, y_pred = _check_targets(y_true, y_pred)
File "/home/personal/Documents/Personal/Documents/aqora/quantum-signals-lob/.venv/lib/python3.10/site-packages/sklearn/metrics/_classification.py", line 103, in _check_targets
check_consistent_length(y_true, y_pred)
File "/home/personal/Documents/Personal/Documents/aqora/quantum-signals-lob/.venv/lib/python3.10/site-packages/sklearn/utils/validation.py", line 457, in check_consistent_length
raise ValueError(
ValueError: Found input variables with inconsistent numbers of samples: [1, 32]
[2 FAIL] Evaluation error
Failed to run pipeline ERROR Oh no! Failed to run pipeline
To try and fix this, you can:
- Check the above error and try againPosted by hasarinduperera •
PS: after the code change, notebook worked fine, I am getting above error, when running
aqora test.To clarify, in initial stage,
aqora test worked, but notebook did not, now after my code change, notebook does work, aqora test is not.Maybe I am doing something wrong or foolish, please forgive me if that's the case.
Posted by antoine •
We do not currently support running submission notebooks directly from Visual Studio Code. Please run your submission using
aqora test for now. We are currently working on integrating aqora test with python notebook runners, but for now you will have to use aqora test.ValueError: Found input variables with inconsistent numbers of samples: [1, 32]
This error is thrown by
sklearn as part of their function check_consistent_length. Have you tried to return different output in your notebook?Posted by hasarinduperera •
Please run your submission using aqora test for now.
Got it, thanks! So basically what I have to do is, experiment with the model and run
aqora test right?Posted by antoine •
Yes exactly 😉 I also highly recommend you subscribe to our newsletter in order to follow our latest development on the platform!
Posted by hasarinduperera •
Sure, I will. Thanks!
Posted by stubbi •
@hasarinduperera thanks for posting!
@antoine can I ask you to take a look at this one? 🙏