Model soups: averaging weights of multiple fine-tuned models improves accuracy without increasing inference time
Paper • 2203.05482 • Published • 8
# Load model directly
from transformers import AutoTokenizer, AutoModelForSequenceClassification
tokenizer = AutoTokenizer.from_pretrained("mllm-dev/gpt2_m_experiment_drug_data_linear_test_new_run")
model = AutoModelForSequenceClassification.from_pretrained("mllm-dev/gpt2_m_experiment_drug_data_linear_test_new_run")Quick Links
tam_test_merge_out_drug_data_dare_linear_test_new_run
This is a merge of pre-trained language models created using mergekit.
Merge Details
Merge Method
This model was merged using the linear merge method.
Models Merged
The following models were included in the merge:
- mllm-dev/gpt2_f_experiment_3_drug_data_new_run
- mllm-dev/gpt2_f_experiment_2_drug_data_new_run
- mllm-dev/gpt2_f_experiment_1_drug_data_new_run
- mllm-dev/gpt2_f_experiment_0_drug_data_new_run
- mllm-dev/gpt2_f_experiment_4_drug_data_new_run
Configuration
The following YAML configuration was used to produce this model:
dtype: float16
merge_method: linear
slices:
- sources:
- layer_range: [0, 12]
model: mllm-dev/gpt2_f_experiment_0_drug_data_new_run
parameters:
weight: 1.0
- layer_range: [0, 12]
model: mllm-dev/gpt2_f_experiment_1_drug_data_new_run
parameters:
weight: 1.0
- layer_range: [0, 12]
model: mllm-dev/gpt2_f_experiment_2_drug_data_new_run
parameters:
weight: 1.0
- layer_range: [0, 12]
model: mllm-dev/gpt2_f_experiment_3_drug_data_new_run
parameters:
weight: 1.0
- layer_range: [0, 12]
model: mllm-dev/gpt2_f_experiment_4_drug_data_new_run
parameters:
weight: 1.0
- Downloads last month
- 2
Model tree for mllm-dev/gpt2_m_experiment_drug_data_linear_test_new_run
Merge model
this model
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="mllm-dev/gpt2_m_experiment_drug_data_linear_test_new_run")