Push model using huggingface_hub.
Browse files- README.md +6 -12
- config.json +4 -4
- training_config.json +32 -4
README.md
CHANGED
|
@@ -1,16 +1,10 @@
|
|
| 1 |
---
|
| 2 |
-
language: en
|
| 3 |
-
license: mit
|
| 4 |
tags:
|
| 5 |
-
-
|
| 6 |
-
-
|
| 7 |
-
- perturbation
|
| 8 |
---
|
| 9 |
|
| 10 |
-
|
| 11 |
-
|
| 12 |
-
|
| 13 |
-
|
| 14 |
-
## Usage
|
| 15 |
-
```python
|
| 16 |
-
model = HFPerturbationTFModel.from_pretrained('weililab/pertTF_virtual_screen_lochness')
|
|
|
|
| 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 |
+
- Code: [More Information Needed]
|
| 9 |
+
- Paper: [More Information Needed]
|
| 10 |
+
- Docs: [More Information Needed]
|
|
|
|
|
|
|
|
|
config.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
{
|
| 2 |
"cell_emb_style": "cls",
|
| 3 |
-
"d_hid":
|
| 4 |
"d_model": 32,
|
| 5 |
"do_dab": false,
|
| 6 |
"do_mvc": true,
|
|
@@ -12,9 +12,9 @@
|
|
| 12 |
"input_emb_style": "continuous",
|
| 13 |
"mvc_decoder_style": "inner product",
|
| 14 |
"n_bins": 0,
|
| 15 |
-
"n_cls":
|
| 16 |
-
"n_pert":
|
| 17 |
-
"n_ps":
|
| 18 |
"nhead": 4,
|
| 19 |
"nlayers": 2,
|
| 20 |
"nlayers_cls": 3,
|
|
|
|
| 1 |
{
|
| 2 |
"cell_emb_style": "cls",
|
| 3 |
+
"d_hid": 32,
|
| 4 |
"d_model": 32,
|
| 5 |
"do_dab": false,
|
| 6 |
"do_mvc": true,
|
|
|
|
| 12 |
"input_emb_style": "continuous",
|
| 13 |
"mvc_decoder_style": "inner product",
|
| 14 |
"n_bins": 0,
|
| 15 |
+
"n_cls": 14,
|
| 16 |
+
"n_pert": 9854,
|
| 17 |
+
"n_ps": 23,
|
| 18 |
"nhead": 4,
|
| 19 |
"nlayers": 2,
|
| 20 |
"nlayers_cls": 3,
|
training_config.json
CHANGED
|
@@ -10,9 +10,6 @@
|
|
| 10 |
"dab_weight": 0.0,
|
| 11 |
"this_weight": 1.0,
|
| 12 |
"next_weight": 0.0,
|
| 13 |
-
"pad_token": "<pad>",
|
| 14 |
-
"mask_token": "<mask>",
|
| 15 |
-
"cls_token": "<cls>",
|
| 16 |
"n_rounds": 10,
|
| 17 |
"next_cell_pred_type": "lochness",
|
| 18 |
"ecs_weight": 1.0,
|
|
@@ -48,5 +45,36 @@
|
|
| 48 |
"fix_nonzero_prop": false,
|
| 49 |
"nonzero_prop": 0.9,
|
| 50 |
"max_seq_len": 3001,
|
| 51 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 52 |
}
|
|
|
|
| 10 |
"dab_weight": 0.0,
|
| 11 |
"this_weight": 1.0,
|
| 12 |
"next_weight": 0.0,
|
|
|
|
|
|
|
|
|
|
| 13 |
"n_rounds": 10,
|
| 14 |
"next_cell_pred_type": "lochness",
|
| 15 |
"ecs_weight": 1.0,
|
|
|
|
| 45 |
"fix_nonzero_prop": false,
|
| 46 |
"nonzero_prop": 0.9,
|
| 47 |
"max_seq_len": 3001,
|
| 48 |
+
"n_pert": 9854,
|
| 49 |
+
"nlayers_pert": 3,
|
| 50 |
+
"n_ps": 23,
|
| 51 |
+
"ntoken": 5014,
|
| 52 |
+
"d_model": 32,
|
| 53 |
+
"nhead": 4,
|
| 54 |
+
"d_hid": 32,
|
| 55 |
+
"nlayers": 2,
|
| 56 |
+
"nlayers_cls": 3,
|
| 57 |
+
"n_cls": 14,
|
| 58 |
+
"vocab": null,
|
| 59 |
+
"dropout": 0.4,
|
| 60 |
+
"pad_token": "<pad>",
|
| 61 |
+
"pad_value": -2,
|
| 62 |
+
"do_mvc": true,
|
| 63 |
+
"do_dab": false,
|
| 64 |
+
"use_batch_labels": false,
|
| 65 |
+
"num_batch_labels": null,
|
| 66 |
+
"domain_spec_batchnorm": false,
|
| 67 |
+
"input_emb_style": "continuous",
|
| 68 |
+
"n_bins": 0,
|
| 69 |
+
"cell_emb_style": "cls",
|
| 70 |
+
"mvc_decoder_style": "inner product",
|
| 71 |
+
"ecs_threshold": 0.7,
|
| 72 |
+
"explicit_zero_prob": true,
|
| 73 |
+
"use_fast_transformer": true,
|
| 74 |
+
"fast_transformer_backend": "flash",
|
| 75 |
+
"pre_norm": false,
|
| 76 |
+
"pred_lochness_next": 100,
|
| 77 |
+
"ps_decoder2_nlayer": 5,
|
| 78 |
+
"pert_pad_id": null,
|
| 79 |
+
"pert_dim": null
|
| 80 |
}
|