library_name: transformers
tags: []
Model Card for Model ID
Model Details
Model Description
I fine-tuned DeBERTa v3 large on answerability of SQaD v2.
Done here: https://colab.research.google.com/drive/1xAA4D3VkbIXYeyIzn5-PE8Xa1miz9uwq#scrollTo=4G7kLtQiFF7Q
import transformers
tokenized_datasets.set_format('torch')
data_collator = transformers.DataCollatorWithPadding(tokenizer=tokenizer)
# Training arguments
training_args = TrainingArguments(
run_name=NOTEBOOK_NAME,
output_dir=NOTEBOOK_NAME,
learning_rate=1e-5, # 3e-5 seemed bad with deepset/roberta-base-squad2, but not sure...
per_device_train_batch_size=16,
gradient_accumulation_steps=2, # others use bs 16, simulate that (2*real-bs)
weight_decay=0.02,
num_train_epochs=2,
fp16=True, # mixed precision training to speed up training and reduce memory usage
evaluation_strategy="steps",
eval_steps=500, # prev 1000, but now we have gradient_accumulation_steps=2
save_strategy="steps",
save_steps=500,
save_total_limit=1, # only save latest and best model
load_best_model_at_end=True,
metric_for_best_model="f1", # to represent precision-recall tradeoff
greater_is_better=True,
report_to=['wandb'],
logging_steps=500,
push_to_hub=True,
warmup_steps=500, # Add learning rate warm-up
lr_scheduler_type="linear", # Use linear decay
# max_grad_norm=1.0, # Clip gradients
)
# Initialize Trainer
trainer = Trainer(
model=model,
args=training_args,
train_dataset=tokenized_datasets['train'],
eval_dataset=tokenized_datasets['validation'],
tokenizer=tokenizer,
compute_metrics=compute_metrics,
data_collator=data_collator, # make batch shorter if possible
# callbacks=[transformers.EarlyStoppingCallback(early_stopping_patience=3)],
)
trainer.train()
Took around 36gb vram (maybe 38gb). 5538/8144 steps took 1:47:48. Trained on a A100. Cost: about 5€
wandb: https://wandb.ai/stadeltom-com/huggingface/runs/thxte3cl
Achieves 93% f1, 92 % precision, and 94% recall
This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
- Developed by: [More Information Needed]
- Funded by [optional]: [More Information Needed]
- Shared by [optional]: [More Information Needed]
- Model type: [More Information Needed]
- Language(s) (NLP): [More Information Needed]
- License: [More Information Needed]
- Finetuned from model [optional]: [More Information Needed]
Model Sources [optional]
- Repository: [More Information Needed]
- Paper [optional]: [More Information Needed]
- Demo [optional]: [More Information Needed]
Uses
Direct Use
[More Information Needed]
Downstream Use [optional]
[More Information Needed]
Out-of-Scope Use
[More Information Needed]
Bias, Risks, and Limitations
[More Information Needed]
Recommendations
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
Training Details
Training Data
[More Information Needed]
Training Procedure
Preprocessing [optional]
[More Information Needed]
Training Hyperparameters
- Training regime: [More Information Needed]
Speeds, Sizes, Times [optional]
[More Information Needed]
Evaluation
Testing Data, Factors & Metrics
Testing Data
[More Information Needed]
Factors
[More Information Needed]
Metrics
[More Information Needed]
Results
[More Information Needed]
Summary
Model Examination [optional]
[More Information Needed]
Environmental Impact
Carbon emissions can be estimated using the Machine Learning Impact calculator presented in Lacoste et al. (2019).
- Hardware Type: [More Information Needed]
- Hours used: [More Information Needed]
- Cloud Provider: [More Information Needed]
- Compute Region: [More Information Needed]
- Carbon Emitted: [More Information Needed]
Technical Specifications [optional]
Model Architecture and Objective
[More Information Needed]
Compute Infrastructure
[More Information Needed]
Hardware
[More Information Needed]
Software
[More Information Needed]
Citation [optional]
BibTeX:
[More Information Needed]
APA:
[More Information Needed]
Glossary [optional]
[More Information Needed]
More Information [optional]
[More Information Needed]
Model Card Authors [optional]
[More Information Needed]
Model Card Contact
[More Information Needed]