Auro Tripathy

Two API changes for TF 1.0

3

I had to make two API changes for TF 1.0 in the `text_cnn` class ... I'm a TF newbie (use Caffe). Kindly review changes. Are they OK? ``` # self.h_pool...

Use of amp.autocast does not improve performance

1

I'm experimenting with `amp.autocast` (automatic mixed precision) with torch version, '1.7.0a0+7036e91' I find the performance has _not improved_ (but slight degraded, 36.2s for AMP vs 30.4s for FP32 with N=1...

Unable to get an AXIS IP cam to work in the home network

9

Hello @WrathChaos: Thank you profusely for this example app and library. Although I was able to build and download your example app into my iPhone, I have **no** experience in...

Questions on the Focal Loss computation in the SSD Chapter

1

Two questions 1. `alpha_t` is unused below. 2. should we use the following to derive the BCECrossEntropyLoss: > ` pt = torch.log(p) * target.float() + torch.log(1.0 - p) * (1...

Question - Is it possible to apply the A3C RL method to this environment?

2

This is quite nice! There are several A3C PyTorch implementations for Atari. Is it possible to do the same with this Truck environment? Thank you.

In Python 2.7 env, stuck after "test_joints.csv is ready"

2

I agree, this is a Pyhton 3 project. Does anyone know how to move forward with the training with Chainer. If you help me, I may become a Chainer fan...

With the following code, `nvidia-smi nvlink -g 0 -i 0` report zero bytes transmitted/received. Same, if I kick off the benchmarks with `--all_reduce_spec=nccl --variable_update=replicated ` from tensorflow.contrib.nccl import all_sum with...

Can vatic be configured for simple object present/absent responses for classification tasks?

Is there a way to apply index.query first bebore the langchain tools kick-in?

4

I generated an index and added it to tools (see below), yet langchain's ReAct agent ignored it. My index (via `index.query()`) builds a better answer :-( ``` index = GPTSimpleVectorIndex.load_from_disk(index_file)...

Error while running tutorial: TypeError: 'mmap' is an invalid keyword argument for Unpickler()

3

I get this error: `TypeError: 'mmap' is an invalid keyword argument for Unpickler()` while loading the model `model = Transformer.from_folder(model_path, dtype=torch.bfloat16)` Model folder looks fine