Rezaul Karim commited on
Commit
725078e
1 Parent(s): 2ec2d86

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +6 -0
README.md CHANGED
@@ -54,6 +54,7 @@ Use the code below to get started with the model.
54
 
55
  ### Training Data
56
 
 
57
  from transformers import GPT2Tokenizer
58
 
59
  dataset = load_dataset("FinGPT/fingpt-sentiment-train")
@@ -99,9 +100,11 @@ tokenized_datasets['train'] = train_test_split['train']
99
 
100
  small_train_dataset = tokenized_datasets["train"].shuffle(seed=42).select(range(100))
101
  small_eval_dataset = tokenized_datasets["test"].shuffle(seed=42).select(range(100))
 
102
 
103
  ### Fine-tune Procedure
104
 
 
105
  from transformers import GPT2ForSequenceClassification
106
  from transformers import TrainingArguments, Trainer
107
 
@@ -126,6 +129,7 @@ trainer = Trainer(
126
  trainer.train()
127
  trainer.evaluate()
128
  trainer.save_model("fine_tuned_finsetiment_model")
 
129
 
130
  #### Training Hyperparameters
131
 
@@ -139,6 +143,7 @@ trainer.save_model("fine_tuned_finsetiment_model")
139
 
140
  ## Evaluation
141
 
 
142
  import evaluate
143
 
144
  metric = evaluate.load("accuracy")
@@ -148,6 +153,7 @@ def compute_metrics(eval_pred):
148
  predictions = np.argmax(logits, axis=-1)
149
 
150
  return metric.compute(predictions=predictions, references=labels)
 
151
 
152
  ### Results
153
 
 
54
 
55
  ### Training Data
56
 
57
+ ```
58
  from transformers import GPT2Tokenizer
59
 
60
  dataset = load_dataset("FinGPT/fingpt-sentiment-train")
 
100
 
101
  small_train_dataset = tokenized_datasets["train"].shuffle(seed=42).select(range(100))
102
  small_eval_dataset = tokenized_datasets["test"].shuffle(seed=42).select(range(100))
103
+ ```
104
 
105
  ### Fine-tune Procedure
106
 
107
+ ```
108
  from transformers import GPT2ForSequenceClassification
109
  from transformers import TrainingArguments, Trainer
110
 
 
129
  trainer.train()
130
  trainer.evaluate()
131
  trainer.save_model("fine_tuned_finsetiment_model")
132
+ ```
133
 
134
  #### Training Hyperparameters
135
 
 
143
 
144
  ## Evaluation
145
 
146
+ ```
147
  import evaluate
148
 
149
  metric = evaluate.load("accuracy")
 
153
  predictions = np.argmax(logits, axis=-1)
154
 
155
  return metric.compute(predictions=predictions, references=labels)
156
+ ```
157
 
158
  ### Results
159