Spaces:
Sleeping
Sleeping
Upload 17 files
Browse files- .gitattributes +1 -0
- ml_models/.keep +0 -0
- ml_models/labels.json +1 -0
- ml_models/ner_model/config.json +141 -0
- ml_models/ner_model/labels.json +1 -0
- ml_models/ner_model/merges.txt +0 -0
- ml_models/ner_model/optimizer.pt +3 -0
- ml_models/ner_model/pytorch_model.bin +3 -0
- ml_models/ner_model/rng_state.pth +3 -0
- ml_models/ner_model/scaler.pt +3 -0
- ml_models/ner_model/scheduler.pt +3 -0
- ml_models/ner_model/special_tokens_map.json +15 -0
- ml_models/ner_model/tokenizer.json +0 -0
- ml_models/ner_model/tokenizer_config.json +17 -0
- ml_models/ner_model/trainer_state.json +0 -0
- ml_models/ner_model/training_args.bin +3 -0
- ml_models/ner_model/vocab.json +0 -0
- ml_models/sentiment_model/model.ft +3 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
ml_models/sentiment_model/model.ft filter=lfs diff=lfs merge=lfs -text
|
ml_models/.keep
ADDED
|
File without changes
|
ml_models/labels.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
["O", "B-AMENITIES", "I-AMENITIES", "I-CLEANLINESS", "B-CLEANLINESS", "I-COMMUNICATION", "B-COMMUNICATION", "B-CONDITION", "I-CONDITION", "I-CUSTOMER_SERVICE", "B-CUSTOMER_SERVICE", "B-EXTERIOR_LIGHTING", "I-EXTERIOR_LIGHTING", "B-FINANCIAL", "I-FINANCIAL", "B-INTERIOR_LIGHTING", "I-INTERIOR_LIGHTING", "B-INTERNET", "I-INTERNET", "B-LANDSCAPING_GROUNDS", "I-LANDSCAPING_GROUNDS", "B-MAINTENANCE_CLEANLINESS", "I-MAINTENANCE_CLEANLINESS", "I-MAINTENANCE_SERVICE", "B-MAINTENANCE_SERVICE", "B-MAINTENANCE_TIMELINESS", "I-MAINTENANCE_TIMELINESS", "B-MOVE_IN_QUALITY", "I-MOVE_IN_QUALITY", "I-NOISE", "B-NOISE", "B-PACKAGES_MAIL", "I-PACKAGES_MAIL", "B-PARKING", "I-PARKING", "I-PESTS", "B-PESTS", "B-PET_WASTE", "I-PET_WASTE", "I-SECURITY", "B-SECURITY", "B-SMOKE", "I-SMOKE", "B-TRASH", "I-TRASH"]
|
ml_models/ner_model/config.json
ADDED
|
@@ -0,0 +1,141 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "/home/ubuntu/lf_output_V12_locked_warmup/checkpoint-506300",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"LongformerForTokenClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_mode": "longformer",
|
| 7 |
+
"attention_probs_dropout_prob": 0.1,
|
| 8 |
+
"attention_window": [
|
| 9 |
+
128,
|
| 10 |
+
128,
|
| 11 |
+
128,
|
| 12 |
+
128,
|
| 13 |
+
128,
|
| 14 |
+
128,
|
| 15 |
+
128,
|
| 16 |
+
128,
|
| 17 |
+
128,
|
| 18 |
+
128,
|
| 19 |
+
128,
|
| 20 |
+
128
|
| 21 |
+
],
|
| 22 |
+
"bos_token_id": 0,
|
| 23 |
+
"classifier_dropout": null,
|
| 24 |
+
"eos_token_id": 2,
|
| 25 |
+
"finetuning_task": "MRPC",
|
| 26 |
+
"gradient_checkpointing": false,
|
| 27 |
+
"hidden_act": "gelu",
|
| 28 |
+
"hidden_dropout_prob": 0.1,
|
| 29 |
+
"hidden_size": 768,
|
| 30 |
+
"id2label": {
|
| 31 |
+
"0": "LABEL_0",
|
| 32 |
+
"1": "LABEL_1",
|
| 33 |
+
"2": "LABEL_2",
|
| 34 |
+
"3": "LABEL_3",
|
| 35 |
+
"4": "LABEL_4",
|
| 36 |
+
"5": "LABEL_5",
|
| 37 |
+
"6": "LABEL_6",
|
| 38 |
+
"7": "LABEL_7",
|
| 39 |
+
"8": "LABEL_8",
|
| 40 |
+
"9": "LABEL_9",
|
| 41 |
+
"10": "LABEL_10",
|
| 42 |
+
"11": "LABEL_11",
|
| 43 |
+
"12": "LABEL_12",
|
| 44 |
+
"13": "LABEL_13",
|
| 45 |
+
"14": "LABEL_14",
|
| 46 |
+
"15": "LABEL_15",
|
| 47 |
+
"16": "LABEL_16",
|
| 48 |
+
"17": "LABEL_17",
|
| 49 |
+
"18": "LABEL_18",
|
| 50 |
+
"19": "LABEL_19",
|
| 51 |
+
"20": "LABEL_20",
|
| 52 |
+
"21": "LABEL_21",
|
| 53 |
+
"22": "LABEL_22",
|
| 54 |
+
"23": "LABEL_23",
|
| 55 |
+
"24": "LABEL_24",
|
| 56 |
+
"25": "LABEL_25",
|
| 57 |
+
"26": "LABEL_26",
|
| 58 |
+
"27": "LABEL_27",
|
| 59 |
+
"28": "LABEL_28",
|
| 60 |
+
"29": "LABEL_29",
|
| 61 |
+
"30": "LABEL_30",
|
| 62 |
+
"31": "LABEL_31",
|
| 63 |
+
"32": "LABEL_32",
|
| 64 |
+
"33": "LABEL_33",
|
| 65 |
+
"34": "LABEL_34",
|
| 66 |
+
"35": "LABEL_35",
|
| 67 |
+
"36": "LABEL_36",
|
| 68 |
+
"37": "LABEL_37",
|
| 69 |
+
"38": "LABEL_38",
|
| 70 |
+
"39": "LABEL_39",
|
| 71 |
+
"40": "LABEL_40",
|
| 72 |
+
"41": "LABEL_41",
|
| 73 |
+
"42": "LABEL_42",
|
| 74 |
+
"43": "LABEL_43",
|
| 75 |
+
"44": "LABEL_44"
|
| 76 |
+
},
|
| 77 |
+
"ignore_attention_mask": false,
|
| 78 |
+
"initializer_range": 0.02,
|
| 79 |
+
"intermediate_size": 3072,
|
| 80 |
+
"label2id": {
|
| 81 |
+
"LABEL_0": 0,
|
| 82 |
+
"LABEL_1": 1,
|
| 83 |
+
"LABEL_10": 10,
|
| 84 |
+
"LABEL_11": 11,
|
| 85 |
+
"LABEL_12": 12,
|
| 86 |
+
"LABEL_13": 13,
|
| 87 |
+
"LABEL_14": 14,
|
| 88 |
+
"LABEL_15": 15,
|
| 89 |
+
"LABEL_16": 16,
|
| 90 |
+
"LABEL_17": 17,
|
| 91 |
+
"LABEL_18": 18,
|
| 92 |
+
"LABEL_19": 19,
|
| 93 |
+
"LABEL_2": 2,
|
| 94 |
+
"LABEL_20": 20,
|
| 95 |
+
"LABEL_21": 21,
|
| 96 |
+
"LABEL_22": 22,
|
| 97 |
+
"LABEL_23": 23,
|
| 98 |
+
"LABEL_24": 24,
|
| 99 |
+
"LABEL_25": 25,
|
| 100 |
+
"LABEL_26": 26,
|
| 101 |
+
"LABEL_27": 27,
|
| 102 |
+
"LABEL_28": 28,
|
| 103 |
+
"LABEL_29": 29,
|
| 104 |
+
"LABEL_3": 3,
|
| 105 |
+
"LABEL_30": 30,
|
| 106 |
+
"LABEL_31": 31,
|
| 107 |
+
"LABEL_32": 32,
|
| 108 |
+
"LABEL_33": 33,
|
| 109 |
+
"LABEL_34": 34,
|
| 110 |
+
"LABEL_35": 35,
|
| 111 |
+
"LABEL_36": 36,
|
| 112 |
+
"LABEL_37": 37,
|
| 113 |
+
"LABEL_38": 38,
|
| 114 |
+
"LABEL_39": 39,
|
| 115 |
+
"LABEL_4": 4,
|
| 116 |
+
"LABEL_40": 40,
|
| 117 |
+
"LABEL_41": 41,
|
| 118 |
+
"LABEL_42": 42,
|
| 119 |
+
"LABEL_43": 43,
|
| 120 |
+
"LABEL_44": 44,
|
| 121 |
+
"LABEL_5": 5,
|
| 122 |
+
"LABEL_6": 6,
|
| 123 |
+
"LABEL_7": 7,
|
| 124 |
+
"LABEL_8": 8,
|
| 125 |
+
"LABEL_9": 9
|
| 126 |
+
},
|
| 127 |
+
"layer_norm_eps": 1e-05,
|
| 128 |
+
"max_position_embeddings": 4098,
|
| 129 |
+
"model_type": "longformer",
|
| 130 |
+
"num_attention_heads": 12,
|
| 131 |
+
"num_hidden_layers": 12,
|
| 132 |
+
"onnx_export": false,
|
| 133 |
+
"pad_token_id": 1,
|
| 134 |
+
"position_embedding_type": "absolute",
|
| 135 |
+
"sep_token_id": 2,
|
| 136 |
+
"torch_dtype": "float32",
|
| 137 |
+
"transformers_version": "4.21.3",
|
| 138 |
+
"type_vocab_size": 1,
|
| 139 |
+
"use_cache": true,
|
| 140 |
+
"vocab_size": 50265
|
| 141 |
+
}
|
ml_models/ner_model/labels.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
["O", "B-AMENITIES", "I-AMENITIES", "I-CLEANLINESS", "B-CLEANLINESS", "I-COMMUNICATION", "B-COMMUNICATION", "B-CONDITION", "I-CONDITION", "I-CUSTOMER_SERVICE", "B-CUSTOMER_SERVICE", "B-EXTERIOR_LIGHTING", "I-EXTERIOR_LIGHTING", "B-FINANCIAL", "I-FINANCIAL", "B-INTERIOR_LIGHTING", "I-INTERIOR_LIGHTING", "B-INTERNET", "I-INTERNET", "B-LANDSCAPING_GROUNDS", "I-LANDSCAPING_GROUNDS", "B-MAINTENANCE_CLEANLINESS", "I-MAINTENANCE_CLEANLINESS", "I-MAINTENANCE_SERVICE", "B-MAINTENANCE_SERVICE", "B-MAINTENANCE_TIMELINESS", "I-MAINTENANCE_TIMELINESS", "B-MOVE_IN_QUALITY", "I-MOVE_IN_QUALITY", "I-NOISE", "B-NOISE", "B-PACKAGES_MAIL", "I-PACKAGES_MAIL", "B-PARKING", "I-PARKING", "I-PESTS", "B-PESTS", "B-PET_WASTE", "I-PET_WASTE", "I-SECURITY", "B-SECURITY", "B-SMOKE", "I-SMOKE", "B-TRASH", "I-TRASH"]
|
ml_models/ner_model/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
ml_models/ner_model/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1fd280344b53bdb98acde083b8c3212310e8093a022fcbaa1ed217055c3cbdd8
|
| 3 |
+
size 1014853157
|
ml_models/ner_model/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8577164b37b26e7f0475bd7ef5fbe011009429ad5f7fd064c7f8d666bdedf8ee
|
| 3 |
+
size 592506275
|
ml_models/ner_model/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8f49b99e408394983a6396520a64117c507c518b9d2e50fa0b7e2e3b6331ea5f
|
| 3 |
+
size 14503
|
ml_models/ner_model/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5bd2fe60a06e67fd651d6fd8675ff8381d60afac5162054085d150fcf0c2cc32
|
| 3 |
+
size 559
|
ml_models/ner_model/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1cd1764a7740f02dc924cefef1fcc750bf0556c02edf2d7cce755bb64f0e7ae3
|
| 3 |
+
size 623
|
ml_models/ner_model/special_tokens_map.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<s>",
|
| 3 |
+
"cls_token": "<s>",
|
| 4 |
+
"eos_token": "</s>",
|
| 5 |
+
"mask_token": {
|
| 6 |
+
"content": "<mask>",
|
| 7 |
+
"lstrip": true,
|
| 8 |
+
"normalized": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false
|
| 11 |
+
},
|
| 12 |
+
"pad_token": "<pad>",
|
| 13 |
+
"sep_token": "</s>",
|
| 14 |
+
"unk_token": "<unk>"
|
| 15 |
+
}
|
ml_models/ner_model/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
ml_models/ner_model/tokenizer_config.json
ADDED
|
@@ -0,0 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": true,
|
| 3 |
+
"bos_token": "<s>",
|
| 4 |
+
"cls_token": "<s>",
|
| 5 |
+
"eos_token": "</s>",
|
| 6 |
+
"errors": "replace",
|
| 7 |
+
"mask_token": "<mask>",
|
| 8 |
+
"max_length": 512,
|
| 9 |
+
"model_max_length": 4096,
|
| 10 |
+
"name_or_path": "/home/ubuntu/lf_output_V12_locked_warmup/checkpoint-506300",
|
| 11 |
+
"pad_token": "<pad>",
|
| 12 |
+
"sep_token": "</s>",
|
| 13 |
+
"special_tokens_map_file": null,
|
| 14 |
+
"tokenizer_class": "LongformerTokenizer",
|
| 15 |
+
"trim_offsets": true,
|
| 16 |
+
"unk_token": "<unk>"
|
| 17 |
+
}
|
ml_models/ner_model/trainer_state.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
ml_models/ner_model/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:84425ccacedcb7b1424919c4151154e33ea9a5bc0b5cc52e15519f30038b9646
|
| 3 |
+
size 3375
|
ml_models/ner_model/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
ml_models/sentiment_model/model.ft
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c94e6fda34edf3a8328a03c4eea9eff9e0704568bd3d63d8148be2c1838dac37
|
| 3 |
+
size 9583103
|