zama-ai / concrete-ml

Concrete ML: Privacy Preserving ML framework built on top of Concrete, with bindings to traditional ML frameworks.

zama-ai/concrete-ml Issues

LLVM symbolizer error with LogisticRegression example
Updated 14 days ago13
Two consecutive Unsqueeze operations in QAT model throws error at compilation time
Updated 20 days ago2
Implement LogSoftmax, Softmax, ReduceMax
Updated 25 days ago3
installation error
Closed a month ago7
[Question] Discord link in explanation
Closed a month ago2
High accuracy variance during the training with SGDClassifier
Updated a month ago1
Performance Issues
Updated a month ago1
Feature Request: Add support for embedding layers
Updated a month ago4
ValueError: Please either set all three 'ordered_module_input_names', 'ordered_module_output_names' and 'quant_layers_dict' or none of them.
Updated a month ago4
[Question] FHE inference over a single image time
Updated a month ago2
[Question] How exactly the HybridFHE functions
Updated a month ago1
RuntimeError: Can't compile: Cannot find crypto parameters
Updated 2 months ago1
Can not install concrete-ml on Linux system
Closed 2 months ago7
Test Hugging Face Endpoints on Azure
Closed 2 months ago2
Add a fhe or concrete-ml as a tag in HF compiled models
Closed 2 months ago2
Display ‘ValueError: vector::_M_default_append’ a when running FHE inference
Closed 3 months ago3
How to use L1-norm unstructured pruning？
Closed 2 months ago3
Faster TFHE Bootstrapping with Block Binary Keys
Closed 2 months ago1
Comparison over encrypted integers
Closed 2 months ago3
Is there a way for the model to return the encrypted test set after performing the inference?
Closed 3 months ago2
Where can decrypt the prediction results? And who is responsible for decrypting them?
Closed 3 months ago5
how
Closed 3 months ago
After model.compile, does the model exist in plaintext form or in ciphertext?
Closed 3 months ago2
Feature Request: Add support for ScatterElements operator during compilation
Updated 3 months ago18
Feature request: support GlobalAveragePooling
Updated 3 months ago7
AttributeError: 'float' object has not attribute 'astype'
Closed 3 months ago18
Failed to compile concrete-ml models on Ubuntu
Closed 4 months ago8
Enhance the documentation for encrypted training
Closed 4 months ago7
Unknown attributes: count_include_pad when compiling with compile_torch_model
Closed 4 months ago2
Quantized Unfold
Closed 4 months ago1
May I ask how to visualize the graph？
Closed 4 months ago1
How to quantify BatchNorm2d layer？
Closed 5 months ago9
Does this library support running on GPU?
Updated 5 months ago3
Feature Request: support of torch.nn.Unfold function
Closed 6 months ago2
Certain ONNX Operators
Closed 6 months ago2
Docker image is not working on Apple M1/M2 chips
Updated 7 months ago11
Can't emit artifacts
Closed 7 months ago3
Accuracy mismatch between clear and FHE predictions in CIFAR-10 example
Closed 7 months ago4
Can FHE inference run on cuda？
Closed 7 months ago1
What is the difference between a graph and a circuit
Closed 9 months ago4
Model dumping not working for Fully Connected Neural Network on MNIST Example
Closed 10 months ago2
is independent of the actual bit precision limitation?
Closed 10 months ago4
WARNING: high error rate, more details with --display-optimizer-choice？
Closed 10 months ago16
ImportError: cannot import name 'enc_split' from 'utils' in LLM case.
Closed a year ago2
predict() got an unexpected keyword argument 'fhe'
Closed a year ago3
What is the difference between two bit？
Closed a year ago2
When using the cifar-10 data set, the program is killed when it runs to the fhe_compatibility function
Closed a year ago6
How much RAM do I need？
Closed a year ago2
WARNING: The shape inference of onnx.brevitas::Quant type is missing？
Closed a year ago1
What determines how much RAM i need when I run a function ‘fhe_compatibility’
Closed a year ago3