seed93
seed93
So any updates here? We are really looking forward to using ZERO3 for boosting.
because of the difference of BN between training and testing
@Alexjap I use the [pull requests](https://github.com/da03/Attention-OCR/pull/26) code and found this bug. Change `cnn_model = CNN(self.img_data, True) #(not self.forward_only))` to `cnn_model = CNN(self.img_data, not self.forward_only) `
The failed format check is not introduced by this merge request.
rebase done
> Awesome thanks for the followup this looks a lot closer!! > > * What's the maintenance plan for `pydracogltf`? Would you be interested in adding it to the `trimesh`...
Shall we merge this request?
The same to me. Face detection is not good, and failed for most meshes for me. I use 1297 meshes captured by a multi-camera system. The texture image is 6000x6000...
抱歉，时间太久了，我也找不到了
@daerduoCarey double precision also fails in some cases. When theta is set to identity matrix, unit test fails. ``` Dtype *theta = this->blob_theta_->mutable_cpu_data(); caffe_set(this->blob_theta_->count(), Dtype(0), theta); theta[0] += 1; theta[4]...