bstraehle commited on
Commit
10e80e0
1 Parent(s): 69c8058

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +3 -8
app.py CHANGED
@@ -32,10 +32,8 @@ def fine_tune_model(base_model_name, dataset_name):
32
 
33
  print("### Dataset")
34
  print(dataset)
35
- print("### Train example")
36
  print(dataset["train"][:1])
37
- print("### Test example")
38
- print(dataset["test"][:1])
39
  print("###")
40
 
41
  # Load model
@@ -57,10 +55,8 @@ def fine_tune_model(base_model_name, dataset_name):
57
 
58
  print("### Pre-processed dataset")
59
  print(dataset)
60
- print("### Train example")
61
  print(dataset["train"][:1])
62
- print("### Test example")
63
- print(dataset["test"][:1])
64
  print("###")
65
 
66
  # Split dataset into training and validation sets
@@ -138,8 +134,7 @@ def prompt_model(model_name, system_prompt, user_prompt, sql_schema):
138
  def load_model(model_name):
139
  model = AutoModelForCausalLM.from_pretrained(model_name, device_map="auto")
140
  tokenizer = AutoTokenizer.from_pretrained(model_name)
141
- tokenizer.padding_side = "right"
142
- #tokenizer.pad_token = tokenizer.eos_token
143
 
144
  return model, tokenizer
145
 
 
32
 
33
  print("### Dataset")
34
  print(dataset)
35
+ print("### Example")
36
  print(dataset["train"][:1])
 
 
37
  print("###")
38
 
39
  # Load model
 
55
 
56
  print("### Pre-processed dataset")
57
  print(dataset)
58
+ print("### Example")
59
  print(dataset["train"][:1])
 
 
60
  print("###")
61
 
62
  # Split dataset into training and validation sets
 
134
  def load_model(model_name):
135
  model = AutoModelForCausalLM.from_pretrained(model_name, device_map="auto")
136
  tokenizer = AutoTokenizer.from_pretrained(model_name)
137
+ tokenizer.pad_token = tokenizer.eos_token
 
138
 
139
  return model, tokenizer
140