Hugging Face's logo

damgomz
/
ft_bs32_lr7_mlm

Fill-Mask
Transformers
Safetensors
English
albert
text-classification
Inference Endpoints
Model card Files Community
ft_bs32_lr7_mlm / README.md
damgomz's picture
damgomz
Upload README.md with huggingface_hub
6487467 verified
preview code
|
raw
history blame contribute delete
No virus
3.52 kB
metadata
language: en
tags:
  - fill-mask
kwargs:
  timestamp: '2024-05-21T05:33:50'
  project_name: ft_bs32_lr7_mlm_emissions_tracker
  run_id: ea51941f-36d1-40ad-93ce-8070f11b32ff
  duration: 28014.593188285828
  emissions: 0.0169520737389673
  emissions_rate: 6.051158274915012e-7
  cpu_power: 42.5
  gpu_power: 0
  ram_power: 3.75
  cpu_energy: 0.3307273301712342
  gpu_energy: 0
  ram_energy: 0.0291816683419049
  energy_consumed: 0.3599089985131388
  country_name: Switzerland
  country_iso_code: CHE
  region: .nan
  cloud_provider: .nan
  cloud_region: .nan
  os: Linux-5.14.0-70.30.1.el9_0.x86_64-x86_64-with-glibc2.34
  python_version: 3.10.4
  codecarbon_version: 2.3.4
  cpu_count: 2
  cpu_model: Intel(R) Xeon(R) Platinum 8360Y CPU @ 2.40GHz
  gpu_count: .nan
  gpu_model: .nan
  longitude: .nan
  latitude: .nan
  ram_total_size: 10
  tracking_mode: machine
  on_cloud: 'N'
  pue: 1

Environmental Impact (CODE CARBON DEFAULT)

Metric Value
Duration (in seconds) 28014.593188285828
Emissions (Co2eq in kg) 0.0169520737389673
CPU power (W) 42.5
GPU power (W) [No GPU]
RAM power (W) 3.75
CPU energy (kWh) 0.3307273301712342
GPU energy (kWh) [No GPU]
RAM energy (kWh) 0.0291816683419049
Consumed energy (kWh) 0.3599089985131388
Country name Switzerland
Cloud provider nan
Cloud region nan
CPU count 2
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) 0.053928091887450215
Emissions (Co2eq in kg) 0.010972382332078616

Note

20 May 2024

My Config

Config Value
checkpoint damgomz/ThunBERT_bs16_lr5_MLM
model_name ft_bs32_lr7_mlm
sequence_length 400
num_epoch 15
learning_rate 5e-07
batch_size 32
weight_decay 0.0
warm_up_prop 0.0
drop_out_prob 0.1
packing_length 100
train_test_split 0.2
num_steps 81450

Training and Testing steps

Epoch Train Loss Test Loss Accuracy Recall
0 0.680462 0.629265 0.695876 0.819018
1 0.582688 0.540363 0.728277 0.802147
2 0.504317 0.481331 0.773196 0.868098
3 0.446641 0.430399 0.808542 0.889571
4 0.400829 0.396269 0.817378 0.881902
5 0.373893 0.376208 0.826951 0.881902
6 0.354504 0.366698 0.834315 0.895706
7 0.343825 0.356863 0.838733 0.849693
8 0.336049 0.356482 0.844624 0.901840
9 0.329104 0.349773 0.852725 0.892638
10 0.323361 0.346467 0.850515 0.880368
11 0.316434 0.344817 0.854934 0.880368
12 0.309111 0.343348 0.857143 0.886503
13 0.304864 0.341717 0.855670 0.878834
14 0.299619 0.344598 0.854934 0.897239