Hugging Face's logo

Scottie201
/
seq2seq_finetuned

Safetensors
English
bart
Model card Files Community
seq2seq_finetuned
File size: 968 Bytes 
105f4ae
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
 
datasets:
- name: Scottie201/seq2seq_data
  split: train
- name: Scottie201/seq2seq_data
  split: validation
decision:
  continue_training: false
  reason: Model performance has not improved enough to warrant continued training.
  suggestions:
  - Consider adjusting the learning rate.
  - Review training data for any inconsistencies.
model:
  architecture: seq2seq
  base_model: facebook/bart-large
  batch_size: 4
  epochs: 3
  learning_rate: 3e-5
  output_dir: /workspaces/wakin_bake/trained_seq2seq
resources:
  device_map: auto
  torch_dtype: bfloat16
  use_gpu: true
tokenization:
  max_length: 512
  padding: max_length
  source_column: input_text
  target_column: target_text
  truncation: true
training_args:
  eval_strategy: epoch
  generation_max_length: 150
  generation_num_beams: 5
  hub_model_id: Scottie201/seq2seq_finetuned
  logging_dir: ./logs
  logging_steps: 1
  push_to_hub: true
  save_steps: 350
  save_total_limit: 1
training_continue: false