Environmental Impact (CODE CARBON DEFAULT)
Metric | Value |
---|---|
Duration (in seconds) | 210553.2843556404 |
Emissions (Co2eq in kg) | 0.2720349058598398 |
CPU power (W) | 42.5 |
GPU power (W) | [No GPU] |
RAM power (W) | 56.25 |
CPU energy (kWh) | 2.4856944445676237 |
GPU energy (kWh) | [No GPU] |
RAM energy (kWh) | 3.289870954465132 |
Consumed energy (kWh) | 5.775565399032714 |
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) | 0.4053150723846078 |
Emissions (Co2eq in kg) | 0.08246670303929249 |
Note
30 April 2024
My Config
Config | Value |
---|---|
checkpoint | albert-base-v2 |
model_name | BERTrand_bs64_lr5 |
sequence_length | 400 |
num_epoch | 12 |
learning_rate | 5e-05 |
batch_size | 64 |
weight_decay | 0.0 |
warm_up_prop | 0 |
drop_out_prob | 0.1 |
packing_length | 100 |
train_test_split | 0.2 |
num_steps | 3148 |
Training and Testing steps
Epoch | Train Loss | Test Loss |
---|---|---|
0.0 | 15.402081 | 13.817045 |
0.5 | 8.054414 | 7.829068 |
1.0 | 4.816239 | 3.114260 |
1.5 | 2.206430 | 2.955595 |
2.0 | 2.189819 | 2.872115 |
2.5 | 2.418134 | 2.865437 |
3.0 | 2.349051 | 2.810524 |
3.5 | 2.102283 | 2.820134 |
4.0 | 1.907061 | 2.957294 |
4.5 | 2.326205 | 2.785392 |
5.0 | 2.257292 | 2.737638 |
5.5 | 2.127350 | 2.733068 |
6.0 | 1.883285 | 2.774372 |
6.5 | 2.100682 | 2.667502 |
7.0 | 2.194973 | 2.628296 |
7.5 | 2.163919 | 2.643665 |
8.0 | 1.850441 | 2.637510 |
8.5 | 1.968181 | 2.632833 |
9.0 | 2.121989 | 2.625116 |
9.5 | 2.136497 | 2.646418 |
10.0 | 1.891819 | 2.655790 |
10.5 | 1.822331 | 2.596789 |
11.0 | 2.116980 | 2.584595 |
11.5 | 2.101184 | 2.588414 |
12.0 | 1.961101 | 2.560519 |
- Downloads last month
- 2
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.