Environmental Impact (CODE CARBON DEFAULT)
Metric | Value |
---|---|
Duration (in seconds) | 211015.4645042419 |
Emissions (Co2eq in kg) | 0.2208665367559387 |
CPU power (W) | 42.5 |
GPU power (W) | [No GPU] |
RAM power (W) | 37.5 |
CPU energy (kWh) | 2.491150588653827 |
GPU energy (kWh) | [No GPU] |
RAM energy (kWh) | 2.1980606118713815 |
Consumed energy (kWh) | 4.689211200525229 |
Country name | Switzerland |
Cloud provider | nan |
Cloud region | nan |
CPU count | 4 |
CPU model | Intel(R) Xeon(R) Platinum 8360Y CPU @ 2.40GHz |
GPU count | nan |
GPU model | nan |
Environmental Impact (for one core)
Metric | Value |
---|---|
CPU energy (kWh) | 0.40620476917066567 |
Emissions (Co2eq in kg) | 0.08264772359749474 |
Note
30 April 2024
My Config
Config | Value |
---|---|
checkpoint | albert-base-v2 |
model_name | BERTrand_bs32_lr6 |
sequence_length | 400 |
num_epoch | 12 |
learning_rate | 5e-06 |
batch_size | 32 |
weight_decay | 0.0 |
warm_up_prop | 0 |
drop_out_prob | 0.1 |
packing_length | 100 |
train_test_split | 0.2 |
num_steps | 6318 |
Training and Testing steps
Epoch | Train Loss | Test Loss |
---|---|---|
0.0 | 15.603048 | 15.109937 |
0.5 | 8.715844 | 8.071290 |
1.0 | 7.608879 | 8.114126 |
1.5 | 7.407612 | 7.914163 |
2.0 | 7.323461 | 7.774658 |
2.5 | 7.248362 | 7.696718 |
3.0 | 7.101276 | 7.856242 |
3.5 | 7.134161 | 7.617901 |
4.0 | 7.105548 | 7.837306 |
4.5 | 7.221799 | 7.653854 |
5.0 | 7.047156 | 7.659136 |
5.5 | 7.080983 | 7.554190 |
6.0 | 7.083629 | 7.670907 |
6.5 | 7.180606 | 7.623875 |
7.0 | 7.036574 | 7.571451 |
7.5 | 7.037596 | 7.550659 |
8.0 | 7.082738 | 7.634689 |
8.5 | 7.136363 | 7.576325 |
9.0 | 7.046428 | 7.594891 |
9.5 | 7.022868 | 7.588534 |
10.0 | 7.075124 | 7.532026 |
10.5 | 7.078401 | 7.519065 |
11.0 | 7.109886 | 7.550544 |
11.5 | 7.002302 | 7.586807 |
12.0 | 7.054321 | 7.544724 |
- Downloads last month
- 15
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.