GraphNNSweep4 | BehaviorCloning Workspace – Weights & Biases

Skip to main content

Batu's group workspace

Group: GraphNNSweep4

1-20

of 45

Tags

Notes

Author

State

Finished

Start time

December 1st, 2021 4:42:10 AM

Runtime

4m 21s

Tracked hours

4m 17s

Run path

batu/BehaviorCloning/1dghg92c

OS

Windows-10-10.0.19041-SP0

Python version

3.7.10

Git repository

git clone https://github.com/batu/RLAgency.git

Git state

git checkout -b "dropoutTrue" 6e4e3bb5250b23f7cc217320753f7a5bca39a6ab

Command

BehaviorCloning.ipynb

System Hardware

CPU count	16
GPU count	1
GPU type	NVIDIA GeForce RTX 3070 Laptop GPU

W&B CLI Version

0.12.6

Group

Config parameters are your model's inputs. Learn more

▶
Config parameters:{} 15 keys
- activation:
  "torch.nn.functional.relu"
- activation_first_layer:
  false
- add_node_id:
  true
- aggregation_method:
  "node"
- batch_size:
  128
- dropout:
  true
- edge_type:
  "goal_to_agent"
- hidden_dim:
  16
- learning_rate:
  0.01
- linear_input:
  true
- linear_output:
  true
- model_type:
  "__main__.GCNLayer"
- num_layers:
  3
- residual:
  false
- self_loop:
  true

Summary metrics are your model's outputs. Learn more

▶
Summary metrics:{} 10 keys
- epoch:
  19
- Mean of Gradients:
  0.006964176893234253
- ▶
  Succes-Fail Ending Position:{} 7 keys
- ▶
  Succes-Fail Starting Position:{} 7 keys
- Success Rate:
  0.916
- train_loss_epoch:
  0.10162505507469176
- train_loss_step:
  0.124915212392807
- trainer/global_step:
  3,519
- val_loss_epoch:
  0.07481145113706589
- val_loss_step:
  0.05551665276288986

This run produced these artifacts as outputs. Total: 2. Learn more

source-1dghg92c:v0

run-1dghg92c-history:v0

Loading...