Add training checkpoints (safetensors)
Browse files- checkpoints/epoch_001/config.json +22 -0
- checkpoints/epoch_001/loss.safetensors +3 -0
- checkpoints/epoch_001/model.safetensors +3 -0
- checkpoints/epoch_001/training_state.pt +3 -0
- checkpoints/epoch_002/config.json +22 -0
- checkpoints/epoch_002/loss.safetensors +3 -0
- checkpoints/epoch_002/model.safetensors +3 -0
- checkpoints/epoch_002/training_state.pt +3 -0
- checkpoints/epoch_003/config.json +22 -0
- checkpoints/epoch_003/loss.safetensors +3 -0
- checkpoints/epoch_003/model.safetensors +3 -0
- checkpoints/epoch_003/training_state.pt +3 -0
- checkpoints/final/aligner_audio.safetensors +3 -0
- checkpoints/final/aligner_code.safetensors +3 -0
- checkpoints/final/aligner_image.safetensors +3 -0
- checkpoints/final/aligner_protein.safetensors +3 -0
- checkpoints/final/config.json +22 -0
- checkpoints/final/loss.safetensors +3 -0
- checkpoints/final/model.safetensors +3 -0
- checkpoints/final/training_state.pt +3 -0
checkpoints/epoch_001/config.json
ADDED
|
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"d_model": 1024,
|
| 3 |
+
"max_text_len": 32,
|
| 4 |
+
"batch_size": 256,
|
| 5 |
+
"eval_batch_size": 512,
|
| 6 |
+
"train_num_workers": 4,
|
| 7 |
+
"eval_num_workers": 2,
|
| 8 |
+
"epochs": 3,
|
| 9 |
+
"lr": 0.0003,
|
| 10 |
+
"weight_decay": 0.01,
|
| 11 |
+
"min_lr": 1e-06,
|
| 12 |
+
"grad_clip": 1.0,
|
| 13 |
+
"align_samples": 5000,
|
| 14 |
+
"whiten_procrustes": true,
|
| 15 |
+
"enforce_rotation_only": false,
|
| 16 |
+
"pw": 0.1,
|
| 17 |
+
"aw": 0.05,
|
| 18 |
+
"checkpoint_dir": "/home/claude/bertenstein_checkpoints",
|
| 19 |
+
"tensorboard_dir": "/home/claude/bertenstein_tb",
|
| 20 |
+
"save_every_epoch": true,
|
| 21 |
+
"log_every_n_steps": 10
|
| 22 |
+
}
|
checkpoints/epoch_001/loss.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ad9993e65ae5aa21ebb3335eda17110e2388f9794676c68c5e844a27b416a51d
|
| 3 |
+
size 76
|
checkpoints/epoch_001/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:60cda49cccf29309156f7635e65fe0188495920b817d03f0224807eb6c08128e
|
| 3 |
+
size 230182176
|
checkpoints/epoch_001/training_state.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2dc0ce3d948c17fa5a89d9b677a3a1acd61572b4c46fed4a788254b87201717d
|
| 3 |
+
size 332454613
|
checkpoints/epoch_002/config.json
ADDED
|
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"d_model": 1024,
|
| 3 |
+
"max_text_len": 32,
|
| 4 |
+
"batch_size": 256,
|
| 5 |
+
"eval_batch_size": 512,
|
| 6 |
+
"train_num_workers": 4,
|
| 7 |
+
"eval_num_workers": 2,
|
| 8 |
+
"epochs": 3,
|
| 9 |
+
"lr": 0.0003,
|
| 10 |
+
"weight_decay": 0.01,
|
| 11 |
+
"min_lr": 1e-06,
|
| 12 |
+
"grad_clip": 1.0,
|
| 13 |
+
"align_samples": 5000,
|
| 14 |
+
"whiten_procrustes": true,
|
| 15 |
+
"enforce_rotation_only": false,
|
| 16 |
+
"pw": 0.1,
|
| 17 |
+
"aw": 0.05,
|
| 18 |
+
"checkpoint_dir": "/home/claude/bertenstein_checkpoints",
|
| 19 |
+
"tensorboard_dir": "/home/claude/bertenstein_tb",
|
| 20 |
+
"save_every_epoch": true,
|
| 21 |
+
"log_every_n_steps": 10
|
| 22 |
+
}
|
checkpoints/epoch_002/loss.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8881ed2a12bd03d7833e84990c28f755c750f3224c75adbf8e5525b8699a45ca
|
| 3 |
+
size 76
|
checkpoints/epoch_002/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e6f008163f9da4d974434b3e48806f95e924fc763c0658e13a51e71602cb9e45
|
| 3 |
+
size 230182176
|
checkpoints/epoch_002/training_state.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fbe81417d29f49c7d68f2dd9e6435232c52cbd01ab353509921b8299a7733e7d
|
| 3 |
+
size 332454613
|
checkpoints/epoch_003/config.json
ADDED
|
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"d_model": 1024,
|
| 3 |
+
"max_text_len": 32,
|
| 4 |
+
"batch_size": 256,
|
| 5 |
+
"eval_batch_size": 512,
|
| 6 |
+
"train_num_workers": 4,
|
| 7 |
+
"eval_num_workers": 2,
|
| 8 |
+
"epochs": 3,
|
| 9 |
+
"lr": 0.0003,
|
| 10 |
+
"weight_decay": 0.01,
|
| 11 |
+
"min_lr": 1e-06,
|
| 12 |
+
"grad_clip": 1.0,
|
| 13 |
+
"align_samples": 5000,
|
| 14 |
+
"whiten_procrustes": true,
|
| 15 |
+
"enforce_rotation_only": false,
|
| 16 |
+
"pw": 0.1,
|
| 17 |
+
"aw": 0.05,
|
| 18 |
+
"checkpoint_dir": "/home/claude/bertenstein_checkpoints",
|
| 19 |
+
"tensorboard_dir": "/home/claude/bertenstein_tb",
|
| 20 |
+
"save_every_epoch": true,
|
| 21 |
+
"log_every_n_steps": 10
|
| 22 |
+
}
|
checkpoints/epoch_003/loss.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:43176f6e043e413c2d38ce1541b4bae5fbf59b3d11943441650cdafeb0d9cb53
|
| 3 |
+
size 76
|
checkpoints/epoch_003/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6879a147e1d52387b86c44911f14a41eb7516d727af15fdf8f405c9ba5886a3d
|
| 3 |
+
size 230182176
|
checkpoints/epoch_003/training_state.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:53f501a892ab5fdc0014a7dbc17ed51a32af1297af5c9dc0015efc70af85d132
|
| 3 |
+
size 332454613
|
checkpoints/final/aligner_audio.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:43ded5c16528e0d4f7c78e416c888699b590d704786d030746a6c42c3c8e29a0
|
| 3 |
+
size 17831328
|
checkpoints/final/aligner_code.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:feb9414d4979e5e39e42bc24925b5277b6d4b5743c4197935c75fd9e7793b51b
|
| 3 |
+
size 15732128
|
checkpoints/final/aligner_image.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3ebf167db9f3adb3adbdc32c51c1c2fcbd7d09cea7f4d01d9504aba7ef3d6f10
|
| 3 |
+
size 12587344
|
checkpoints/final/aligner_protein.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:370ec7fa43c69ab5eda8d68e16c500d62b82d9b90580b3c932c68443375a616e
|
| 3 |
+
size 17831328
|
checkpoints/final/config.json
ADDED
|
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"d_model": 1024,
|
| 3 |
+
"max_text_len": 32,
|
| 4 |
+
"batch_size": 256,
|
| 5 |
+
"eval_batch_size": 512,
|
| 6 |
+
"train_num_workers": 4,
|
| 7 |
+
"eval_num_workers": 2,
|
| 8 |
+
"epochs": 3,
|
| 9 |
+
"lr": 0.0003,
|
| 10 |
+
"weight_decay": 0.01,
|
| 11 |
+
"min_lr": 1e-06,
|
| 12 |
+
"grad_clip": 1.0,
|
| 13 |
+
"align_samples": 5000,
|
| 14 |
+
"whiten_procrustes": true,
|
| 15 |
+
"enforce_rotation_only": false,
|
| 16 |
+
"pw": 0.1,
|
| 17 |
+
"aw": 0.05,
|
| 18 |
+
"checkpoint_dir": "/home/claude/bertenstein_checkpoints",
|
| 19 |
+
"tensorboard_dir": "/home/claude/bertenstein_tb",
|
| 20 |
+
"save_every_epoch": true,
|
| 21 |
+
"log_every_n_steps": 10
|
| 22 |
+
}
|
checkpoints/final/loss.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:43176f6e043e413c2d38ce1541b4bae5fbf59b3d11943441650cdafeb0d9cb53
|
| 3 |
+
size 76
|
checkpoints/final/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6879a147e1d52387b86c44911f14a41eb7516d727af15fdf8f405c9ba5886a3d
|
| 3 |
+
size 230182176
|
checkpoints/final/training_state.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:53f501a892ab5fdc0014a7dbc17ed51a32af1297af5c9dc0015efc70af85d132
|
| 3 |
+
size 332454613
|