File tree Expand file tree Collapse file tree 4 files changed +12
-12
lines changed Expand file tree Collapse file tree 4 files changed +12
-12
lines changed Original file line number Diff line number Diff line change 2
2
3
3
# Project metadata
4
4
project_name : olmocr-qwen-vl-training
5
- run_name : qwen2.5-vl-7b-finetune-day2 -json
5
+ run_name : qwen2.5-vl-7b-finetune-day3 -json
6
6
7
7
# Model configuration
8
8
model :
@@ -58,7 +58,7 @@ dataset:
58
58
59
59
# Training configuration
60
60
training :
61
- output_dir : /home/ubuntu /olmocr-trainer/
61
+ output_dir : /weka/oe-data-default/jakep /olmocr-trainer/
62
62
num_train_epochs : 1
63
63
64
64
# Batch size and accumulation
@@ -72,7 +72,7 @@ training:
72
72
73
73
# Learning rate
74
74
learning_rate : 2e-5
75
- lr_scheduler_type : cosine
75
+ lr_scheduler_type : linear
76
76
warmup_ratio : 0.1
77
77
78
78
# Optimization
Original file line number Diff line number Diff line change 2
2
3
3
# Project metadata
4
4
project_name : olmocr-qwen-vl-training
5
- run_name : qwen2.5-vl-7b-finetune-day2 -1280
5
+ run_name : qwen2.5-vl-7b-finetune-day3 -1280
6
6
7
7
# Model configuration
8
8
model :
@@ -58,7 +58,7 @@ dataset:
58
58
59
59
# Training configuration
60
60
training :
61
- output_dir : /home/ubuntu /olmocr-trainer/
61
+ output_dir : /weka/oe-data-default/jakep /olmocr-trainer/
62
62
num_train_epochs : 1
63
63
64
64
# Batch size and accumulation
@@ -72,7 +72,7 @@ training:
72
72
73
73
# Learning rate
74
74
learning_rate : 2e-5
75
- lr_scheduler_type : cosine
75
+ lr_scheduler_type : linear
76
76
warmup_ratio : 0.1
77
77
78
78
# Optimization
Original file line number Diff line number Diff line change 2
2
3
3
# Project metadata
4
4
project_name : olmocr-qwen-vl-training
5
- run_name : qwen2.5-vl-7b-finetune-day2 -1280-noanchor
5
+ run_name : qwen2.5-vl-7b-finetune-day3 -1280-noanchor
6
6
7
7
# Model configuration
8
8
model :
@@ -58,7 +58,7 @@ dataset:
58
58
59
59
# Training configuration
60
60
training :
61
- output_dir : /home/ubuntu /olmocr-trainer/
61
+ output_dir : /weka/oe-data-default/jakep /olmocr-trainer/
62
62
num_train_epochs : 1
63
63
64
64
# Batch size and accumulation
@@ -72,7 +72,7 @@ training:
72
72
73
73
# Learning rate
74
74
learning_rate : 2e-5
75
- lr_scheduler_type : cosine
75
+ lr_scheduler_type : linear
76
76
warmup_ratio : 0.1
77
77
78
78
# Optimization
Original file line number Diff line number Diff line change 2
2
3
3
# Project metadata
4
4
project_name : olmocr-qwen-vl-training
5
- run_name : qwen2.5-vl-7b-finetune-day2 -1600
5
+ run_name : qwen2.5-vl-7b-finetune-day3 -1600
6
6
7
7
# Model configuration
8
8
model :
@@ -58,7 +58,7 @@ dataset:
58
58
59
59
# Training configuration
60
60
training :
61
- output_dir : /home/ubuntu /olmocr-trainer/
61
+ output_dir : /weka/oe-data-default/jakep /olmocr-trainer/
62
62
num_train_epochs : 1
63
63
64
64
# Batch size and accumulation
@@ -72,7 +72,7 @@ training:
72
72
73
73
# Learning rate
74
74
learning_rate : 2e-5
75
- lr_scheduler_type : cosine
75
+ lr_scheduler_type : linear
76
76
warmup_ratio : 0.1
77
77
78
78
# Optimization
You can’t perform that action at this time.
0 commit comments