Push model using huggingface_hub.
Browse files- README.md +9 -0
- config.json +62 -0
- model.safetensors +3 -0
README.md
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
tags:
|
| 3 |
+
- model_hub_mixin
|
| 4 |
+
- pytorch_model_hub_mixin
|
| 5 |
+
---
|
| 6 |
+
|
| 7 |
+
This model has been pushed to the Hub using the [PytorchModelHubMixin](https://huggingface.co/docs/huggingface_hub/package_reference/mixins#huggingface_hub.PyTorchModelHubMixin) integration:
|
| 8 |
+
- Library: [More Information Needed]
|
| 9 |
+
- Docs: [More Information Needed]
|
config.json
ADDED
|
@@ -0,0 +1,62 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"adam_beta1": 0.9,
|
| 3 |
+
"adam_beta2": 0.999,
|
| 4 |
+
"adam_weight_decay": 0.01,
|
| 5 |
+
"amp": false,
|
| 6 |
+
"class_conditional": false,
|
| 7 |
+
"class_unconditional_prob": 0.1,
|
| 8 |
+
"clip_grad_norm": 1.0,
|
| 9 |
+
"dataset_name": "roc",
|
| 10 |
+
"dim_ae": 64,
|
| 11 |
+
"disable_dropout": false,
|
| 12 |
+
"dropout": 0.1,
|
| 13 |
+
"ema_decay": 0.9999,
|
| 14 |
+
"ema_update_every": 1,
|
| 15 |
+
"enc_dec_model": "facebook/bart-base",
|
| 16 |
+
"eval": false,
|
| 17 |
+
"eval_batch_size": 32,
|
| 18 |
+
"eval_every": 1000,
|
| 19 |
+
"eval_test": false,
|
| 20 |
+
"gradient_accumulation_steps": 1,
|
| 21 |
+
"init_path": null,
|
| 22 |
+
"l2_normalize_latents": true,
|
| 23 |
+
"latent_dim": 64,
|
| 24 |
+
"latent_model_path": "saved_latent_models/roc/2024-11-24_09-55-03",
|
| 25 |
+
"learning_rate": 0.0001,
|
| 26 |
+
"lm_mode": "freeze",
|
| 27 |
+
"loss_type": "l2",
|
| 28 |
+
"lr_schedule": "linear",
|
| 29 |
+
"lr_warmup_steps": 1000,
|
| 30 |
+
"max_seq_len": 64,
|
| 31 |
+
"mixed_precision": "no",
|
| 32 |
+
"normalize_latent": false,
|
| 33 |
+
"num_decoder_latents": 32,
|
| 34 |
+
"num_dense_connections": 3,
|
| 35 |
+
"num_devices": 1,
|
| 36 |
+
"num_encoder_latents": 32,
|
| 37 |
+
"num_layers": 3,
|
| 38 |
+
"num_samples": 1000,
|
| 39 |
+
"num_train_steps": 50000,
|
| 40 |
+
"objective": "pred_v",
|
| 41 |
+
"optimizer": "adamw",
|
| 42 |
+
"output_dir": "saved_latent_models/roc/2024-11-24_09-55-03",
|
| 43 |
+
"resume_dir": null,
|
| 44 |
+
"resume_training": false,
|
| 45 |
+
"sampler": "ddpm",
|
| 46 |
+
"sampling_schedule": null,
|
| 47 |
+
"sampling_timesteps": 250,
|
| 48 |
+
"save_and_sample_every": 5000,
|
| 49 |
+
"save_dir": "saved_latent_models",
|
| 50 |
+
"scale": 1.0,
|
| 51 |
+
"scale_shift": true,
|
| 52 |
+
"self_condition": true,
|
| 53 |
+
"seq2seq_candidates": 5,
|
| 54 |
+
"seq2seq_unconditional_prob": 0.1,
|
| 55 |
+
"train_batch_size": 256,
|
| 56 |
+
"train_prob_self_cond": 0.5,
|
| 57 |
+
"train_schedule": "cosine",
|
| 58 |
+
"trainable_params": 187928960,
|
| 59 |
+
"tx_depth": 12,
|
| 60 |
+
"tx_dim": 768,
|
| 61 |
+
"wandb_name": "bart-roc-l2norm-test-32-64"
|
| 62 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3905c3d8fd3f3e2b65ac310daa62f3ad2c0eea836a161b6e0ca09ff7d0ce3d7d
|
| 3 |
+
size 1412158832
|