Spaces:
Runtime error
Runtime error
stakelovelace
commited on
Commit
•
2069fff
1
Parent(s):
339b8e7
commit from tesla
Browse files
app.py
CHANGED
@@ -60,6 +60,12 @@ def train_model(model, tokenizer, data, device):
|
|
60 |
|
61 |
trainer.train()
|
62 |
|
|
|
|
|
|
|
|
|
|
|
|
|
63 |
# Perform any remaining steps such as logging, saving, etc.
|
64 |
trainer.save_model()
|
65 |
|
@@ -70,8 +76,8 @@ def main(api_name, base_url):
|
|
70 |
# Load the configuration for a specific model
|
71 |
config = AutoConfig.from_pretrained('google/codegemma-2b')
|
72 |
# Update the activation function
|
73 |
-
config.hidden_act = '' # Set to use approximate GeLU gelu_pytorch_tanh
|
74 |
-
config.hidden_activation = '
|
75 |
|
76 |
model = AutoModelForCausalLM.from_pretrained('google/codegemma-2b', is_decoder=True)
|
77 |
#model = BertLMHeadModel.from_pretrained('google/codegemma-2b', is_decoder=True)
|
@@ -80,7 +86,6 @@ def main(api_name, base_url):
|
|
80 |
model.to(device) # Move model to the appropriate device
|
81 |
|
82 |
train_model(model, tokenizer, data, device)
|
83 |
-
|
84 |
|
85 |
model.save_pretrained("./fine_tuned_model")
|
86 |
tokenizer.save_pretrained("./fine_tuned_model")
|
|
|
60 |
|
61 |
trainer.train()
|
62 |
|
63 |
+
# Optionally clear cache if using GPU or MPS
|
64 |
+
if torch.cuda.is_available():
|
65 |
+
torch.cuda.empty_cache()
|
66 |
+
elif torch.has_mps:
|
67 |
+
torch.mps.empty_cache()
|
68 |
+
|
69 |
# Perform any remaining steps such as logging, saving, etc.
|
70 |
trainer.save_model()
|
71 |
|
|
|
76 |
# Load the configuration for a specific model
|
77 |
config = AutoConfig.from_pretrained('google/codegemma-2b')
|
78 |
# Update the activation function
|
79 |
+
# config.hidden_act = '' # Set to use approximate GeLU gelu_pytorch_tanh
|
80 |
+
config.hidden_activation = 'gelu_pytorch_tanh' # Set to use GeLU
|
81 |
|
82 |
model = AutoModelForCausalLM.from_pretrained('google/codegemma-2b', is_decoder=True)
|
83 |
#model = BertLMHeadModel.from_pretrained('google/codegemma-2b', is_decoder=True)
|
|
|
86 |
model.to(device) # Move model to the appropriate device
|
87 |
|
88 |
train_model(model, tokenizer, data, device)
|
|
|
89 |
|
90 |
model.save_pretrained("./fine_tuned_model")
|
91 |
tokenizer.save_pretrained("./fine_tuned_model")
|