Downloads · 30 days
0
0% of all-time downloads
bert-tiny_mnli_204ea4ae32
jedidiahkkh/bert-tiny_mnli_204ea4ae32
bert-tiny_mnli_204ea4ae32 is a machine learning model from jedidiahkkh. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
- completehash: 204ea280624a34e057df76d645c4ae32 - shorthash: 204ea4ae32
All-time downloads
29
Public
Repo size
124 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.json713 KB · 43%
At a glance
- Access
- Public
- Created
- May 9, 2024
- Updated
- May 9, 2024
- SHA
- 2703d951
Model card
From the Hugging Face model README
Run info
- complete_hash: 204ea280624a34e057df76d645c4ae32
- short_hash: 204ea4ae32
Configuration
data:
batch_size: 32
data_seed: 42
drop_last: false
eval_batch_size: 128
max_length: 512
multiprocessing_context: null
num_workers: 32
persistent_workers: false
pin_memory: true
replacement: false
shuffle: true
dataset: mnli
estimator:
accelerator: gpu
convert_to_bettertransformer: false
deterministic: true
precision: bf16-true
tf32_mode: high
fit:
enable_progress_bar: true
limit_train_batches: null
log_interval: 100
max_epochs: 6
min_epochs: null
optimizer_kwargs:
init_kwargs:
fused: true
lr: 0.0003
name: adamw
scheduler_kwargs:
name: constant_schedule_with_warmup
num_warmup_steps: 2000
model:
base_model: bert-tiny
name: bert-tiny
revision: null
seed: 42
seed: 42
- Created
- May 9, 2024
- Updated
- May 9, 2024