Upload folder using huggingface_hub
Browse files- gliner_config.json +9 -7
- pytorch_model.bin +1 -1
- rng_state.pth +3 -0
gliner_config.json
CHANGED
@@ -43,6 +43,7 @@
|
|
43 |
"max_length": 20,
|
44 |
"max_position_embeddings": 4096,
|
45 |
"min_length": 0,
|
|
|
46 |
"model_type": "llama",
|
47 |
"no_repeat_ngram_size": 0,
|
48 |
"num_attention_heads": 16,
|
@@ -84,17 +85,18 @@
|
|
84 |
"vocab_size": 32002
|
85 |
},
|
86 |
"ent_token": "<<ENT>>",
|
87 |
-
"eval_every":
|
88 |
"fine_tune": true,
|
89 |
"freeze_token_rep": false,
|
90 |
"fuse_layers": false,
|
|
|
91 |
"has_rnn": true,
|
92 |
"hidden_size": 2048,
|
93 |
"label_smoothing": 0.0001,
|
94 |
"labels_encoder": null,
|
95 |
"labels_encoder_config": null,
|
96 |
-
"log_dir": "llama",
|
97 |
-
"loss_alpha": 0.
|
98 |
"loss_gamma": 2,
|
99 |
"loss_reduction": "sum",
|
100 |
"lr_encoder": "8e-6",
|
@@ -108,13 +110,13 @@
|
|
108 |
"model_type": "gliner",
|
109 |
"name": "span level gliner",
|
110 |
"num_post_fusion_layers": 3,
|
111 |
-
"num_steps":
|
112 |
"post_fusion_schema": "",
|
113 |
"prev_path": null,
|
114 |
"random_drop": true,
|
115 |
"root_dir": "gliner_logs",
|
116 |
-
"save_total_limit":
|
117 |
-
"scheduler_type": "
|
118 |
"sep_token": "<<SEP>>",
|
119 |
"shuffle_types": true,
|
120 |
"size_sup": -1,
|
@@ -122,7 +124,7 @@
|
|
122 |
"subtoken_pooling": "first",
|
123 |
"train_batch_size": 8,
|
124 |
"train_data": "data.json",
|
125 |
-
"transformers_version": "4.
|
126 |
"val_data_dir": "none",
|
127 |
"vocab_size": 32002,
|
128 |
"warmup_ratio": 0.1,
|
|
|
43 |
"max_length": 20,
|
44 |
"max_position_embeddings": 4096,
|
45 |
"min_length": 0,
|
46 |
+
"mlp_bias": false,
|
47 |
"model_type": "llama",
|
48 |
"no_repeat_ngram_size": 0,
|
49 |
"num_attention_heads": 16,
|
|
|
85 |
"vocab_size": 32002
|
86 |
},
|
87 |
"ent_token": "<<ENT>>",
|
88 |
+
"eval_every": 1000,
|
89 |
"fine_tune": true,
|
90 |
"freeze_token_rep": false,
|
91 |
"fuse_layers": false,
|
92 |
+
"has_post_transformer": false,
|
93 |
"has_rnn": true,
|
94 |
"hidden_size": 2048,
|
95 |
"label_smoothing": 0.0001,
|
96 |
"labels_encoder": null,
|
97 |
"labels_encoder_config": null,
|
98 |
+
"log_dir": "models/llama",
|
99 |
+
"loss_alpha": 0.75,
|
100 |
"loss_gamma": 2,
|
101 |
"loss_reduction": "sum",
|
102 |
"lr_encoder": "8e-6",
|
|
|
110 |
"model_type": "gliner",
|
111 |
"name": "span level gliner",
|
112 |
"num_post_fusion_layers": 3,
|
113 |
+
"num_steps": 6000,
|
114 |
"post_fusion_schema": "",
|
115 |
"prev_path": null,
|
116 |
"random_drop": true,
|
117 |
"root_dir": "gliner_logs",
|
118 |
+
"save_total_limit": 5,
|
119 |
+
"scheduler_type": "cosine",
|
120 |
"sep_token": "<<SEP>>",
|
121 |
"shuffle_types": true,
|
122 |
"size_sup": -1,
|
|
|
124 |
"subtoken_pooling": "first",
|
125 |
"train_batch_size": 8,
|
126 |
"train_data": "data.json",
|
127 |
+
"transformers_version": "4.41.0",
|
128 |
"val_data_dir": "none",
|
129 |
"vocab_size": 32002,
|
130 |
"warmup_ratio": 0.1,
|
pytorch_model.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 5824576326
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6e7ed28c6280ca0f67b6fd706451fae9ed8ec7feef033b7ffc1d4c598fce714e
|
3 |
size 5824576326
|
rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d70c43962143e6cd0e0aefc33c812d52bbd815bec0182e2cb359410c6a180bb1
|
3 |
+
size 14244
|