Comments (7)
That first module is the context encoder used to generate comparative statistics for the paper. I wasn't planning on uploading that but I can do if you'd still like it?
So I ran the model on the leather subset of MVTec so you should be able to use the MTec Dataset class and load in the model that way - although when instantiating and using the model throughout I've referred to it as "leather", but you should be able to edit those sections and run it ok - the config file I used looks like (changing the noise_fn for simplex for that model):
{
"img_size": [256,256],
"Batch_Size": 1,
"EPOCHS": 3000,
"T": 1000,
"base_channels": 128,
"beta_schedule": "linear",
"channel_mults": "",
"loss-type": "l2",
"loss_weight": "none",
"train_start": true,
"lr": 1e-4,
"random_slice": true,
"sample_distance": 600,
"weight_decay": 0.0,
"save_imgs":false,
"save_vids":true,
"dropout":0,
"attention_resolutions":"16,8",
"num_heads":2,
"num_head_channels":-1,
"noise_fn":"gauss",
"dataset":"leather"
}
Let me know if this helps and get back in touch if not
from anoddpm.
Thank you very much for your advice. Now I can train the model. By the way, Can you explain in detail how to evaluate the model ? I don't know how to use detection.py
from anoddpm.
Firstly, you must ensure the function you want to run is selected at the bottom of the file and linked under the arg file provided.
Since you're running the model on MVTec, the anomalous_metric_calculation function will loop over the leather dataset and record various relevant metrics to a file. This is the primary function used for evaluation.
Alternatively, if you want to reproduce the graphs for example, then the roc_data and graph_data functions sort this however they would need to be adapted for your dataset and remove redundant comparative models.
Let me know whether this helps
from anoddpm.
Closing due to no responses
from anoddpm.
Thank you very much for your advice. Now I can train the model. By the way, Can you explain in detail how to evaluate the model ? I don't know how to use detection.py
Hello, I am also facing the issue with "import Comparitive_models.CE as CE". I checked this thread but couldn't find the uploaded Comparitive_models. Have you found a solution to it?
from anoddpm.
In the file "detection.py," I found "import Comparative_models.CE as CE" on line 466. How I can get this package? Moreover, I want to run your model based on the dataset "MVTec". Would you please advise me on how I can reproduce the result? What the configuration file should look like and the steps to reproduce the result.
Import Comparative_ Models CE as CE, how did you solve it
from anoddpm.
Thank you very much for your advice. Now I can train the model. By the way, Can you explain in detail how to evaluate the model ? I don't know how to use detection.py
Hello, I am also facing the issue with "import Comparitive_models.CE as CE". I checked this thread but couldn't find the uploaded Comparitive_models. Have you found a solution to it?
I don't believe I have access to that file anymore, just delete the references to it as it is just required for comparing models.
from anoddpm.
Related Issues (20)
- some doubts about the principle of the diffusion model HOT 1
- some quesion about detection HOT 5
- some questions about DDPM HOT 1
- Some problems with running detection HOT 2
- Batch Size HOT 3
- Where is the file of args used in the paper? HOT 2
- How Train and Test is split. HOT 2
- comparasion_models HOT 5
- What is a directory raw_cleaned ? HOT 7
- about the training time cost HOT 1
- Some problems about data HOT 1
- Encountered Access Issue of E-mail HOT 1
- Package Versions / Install Requirements? HOT 1
- how to get testData HOT 2
- high CPU usage discovered in training stage HOT 1
- result for MVTec HOT 1
- Asking about "import Comparative_models.CE as CE" HOT 2
- Problem about saving .mp4 files during training process HOT 3
- Access to training args.json used in the paper
- Pre-trained data HOT 1
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from anoddpm.