-
Hi,
First of all thank you provide the project code generously. I'am trying to train the network on BSDS dataset. But the loss can not convergence. Then, I can understand the value of max_iter equals…
-
When I run on large images with sparse objects, I find that there's a lot of unnecessary masking going on.
My hypothesis is that some of the issue is coming from the form of the classification loss…
-
HI,
Is it possible to add tp/sl partial orders after postion has ben opened?
Example:
Postion entry price: 100
Quantity: 100 some coins
I want to add next orders:
Market takeprofit for 50 co…
-
Hey guys, how are you? I'm doing some tests with this bot and I would like to know if you have already achieved any profitable strategy/configuration?
-
Hi,
I wanted to share the way to help with this problem:
> Training seems to suffer from mode collapse (when trained wrong, enhanced image looks like negatively processed image). When Enhanced-O…
hcl14 updated
5 years ago
-
Would be cool to be able to also add a stop market order when you submit your limit orders
rrcdr updated
5 years ago
-
I have trained catSeq model and its performance is as your reported. When I use
`python3 train.py -data data/kp20k/kp20k_separated/rl/ -vocab data/kp20k/kp20k_separated/rl/ -exp_path=exp -exp catSeq_…
-
I try to attack llama-2-13b, and I ran the run_gcg_multiple.sh, the command is
`
export model=llama2_13
export set_name=merged
export CUDA_VISIBLE_DEVICES=7
export batch_size=512
python -…
-
When inputting long sentences, I found the model tended to repeat the ending words over and over again. I trained this model over the blizzard2011 challenge database and both of the transformer and po…
-
Consider the following example: https://diffeq.sciml.ai/stable/analysis/parameter_estimation/#Optimization-Based-ODE-Inference-Examples
If one makes an `L2Loss` there is nothing that stops one from d…