Upload folder using huggingface_hub
Browse files- README.md +83 -93
- config.json +1 -1
- config_sentence_transformers.json +1 -1
- config_setfit.json +3 -3
- heads/domain-router.pkl +2 -2
- heads/materials-classifier.pkl +2 -2
- model_head.pkl +2 -2
README.md
CHANGED
|
@@ -5,89 +5,79 @@ tags:
|
|
| 5 |
- text-classification
|
| 6 |
- generated_from_setfit_trainer
|
| 7 |
widget:
|
| 8 |
-
- text:
|
| 9 |
-
|
| 10 |
-
|
| 11 |
-
|
| 12 |
-
Trim to size, hasElementID:
|
| 13 |
-
|
| 14 |
-
|
| 15 |
-
|
| 16 |
-
|
| 17 |
-
|
| 18 |
-
|
| 19 |
-
|
| 20 |
-
|
| 21 |
-
1
|
| 22 |
-
|
| 23 |
-
|
| 24 |
-
|
| 25 |
-
|
| 26 |
-
|
| 27 |
-
|
| 28 |
-
|
| 29 |
-
|
| 30 |
-
|
| 31 |
-
|
| 32 |
-
|
| 33 |
-
|
| 34 |
-
|
| 35 |
-
|
| 36 |
-
|
| 37 |
-
|
| 38 |
-
|
| 39 |
-
|
| 40 |
-
|
| 41 |
-
|
| 42 |
-
|
| 43 |
-
|
| 44 |
-
|
| 45 |
-
|
| 46 |
-
|
| 47 |
-
|
| 48 |
-
|
| 49 |
-
|
| 50 |
-
|
| 51 |
-
|
| 52 |
-
|
| 53 |
-
|
| 54 |
-
|
| 55 |
-
|
| 56 |
-
|
| 57 |
-
|
| 58 |
-
|
| 59 |
-
|
| 60 |
-
|
| 61 |
-
|
| 62 |
-
|
| 63 |
-
|
| 64 |
-
|
| 65 |
-
|
| 66 |
-
|
| 67 |
-
|
| 68 |
-
|
| 69 |
-
|
| 70 |
-
- text: 'coloursToFaceAndReverseAreSame: Yes, hasAdditionalInformation: LINE 02 REVISION
|
| 71 |
-
IS FOR QUANTITY ONLY. ALL SPEC''S SAME AS LINE 01. PLEASE NOTE SCHEDULE FOR
|
| 72 |
-
THIS PROJECT, hasArtworkDoubleSidedStatus: Double Sided Different, hasCoatingOrSealer:
|
| 73 |
-
Aqueous, hasColourDetails: prints 4/4 - process 4/c + OA Satin AQ (not required
|
| 74 |
-
if printing digitally), hasCreatedDate: 2025-01-11, hasCustomerHomeCountry: United
|
| 75 |
-
States, hasCustomerID: 12231, hasCustomerName: Janssen Pharmaceuticals, Inc(Janssen
|
| 76 |
-
Pharmaceuticals, Inc), hasCutting: Trim to size, hasElementID: 3735612, hasElementTitle:
|
| 77 |
-
cp-118514v9 - J&J WithMe Getting Started Tearpad, hasFinishedSizeHeight: 11,
|
| 78 |
hasFinishedSizeWidth: 8.5, hasFlatSizeHeight: 11, hasFlatSizeWidth: 8.5, hasFscPaperBeenSpecified:
|
| 79 |
-
|
| 80 |
-
Yes, hasMachineFinishingDetails:
|
| 81 |
-
|
| 82 |
-
|
| 83 |
-
|
| 84 |
-
|
| 85 |
-
|
| 86 |
-
|
| 87 |
-
|
| 88 |
-
|
| 89 |
-
Capital Printing Corporation(Capital Printing Corp. - 18444 - HHGSP), hasTotalColours:
|
| 90 |
-
4, hasTotalColoursFace: 4, hasUnitOfMeasure: Inches (in), '
|
| 91 |
metrics:
|
| 92 |
- f1_micro
|
| 93 |
- f1_macro
|
|
@@ -110,22 +100,22 @@ model-index:
|
|
| 110 |
split: test
|
| 111 |
metrics:
|
| 112 |
- type: f1_micro
|
| 113 |
-
value: 0.
|
| 114 |
name: F1_Micro
|
| 115 |
- type: f1_macro
|
| 116 |
-
value: 0.
|
| 117 |
name: F1_Macro
|
| 118 |
- type: f1_weighted
|
| 119 |
-
value: 0.
|
| 120 |
name: F1_Weighted
|
| 121 |
- type: precision
|
| 122 |
-
value: .
|
| 123 |
name: Precision
|
| 124 |
- type: accuracy
|
| 125 |
-
value:
|
| 126 |
name: Accuracy
|
| 127 |
- type: recall
|
| 128 |
-
value: .
|
| 129 |
name: Recall
|
| 130 |
---
|
| 131 |
|
|
@@ -161,7 +151,7 @@ The model has been trained using an efficient few-shot learning technique that i
|
|
| 161 |
### Metrics
|
| 162 |
| Label | F1_Micro | F1_Macro | F1_Weighted | Precision | Accuracy | Recall |
|
| 163 |
|:--------|:---------|:---------|:------------|:----------|:---------|:-------|
|
| 164 |
-
| **all** | 0.0
|
| 165 |
|
| 166 |
## Uses
|
| 167 |
|
|
@@ -181,7 +171,7 @@ from setfit import SetFitModel
|
|
| 181 |
# Download from the 🤗 Hub
|
| 182 |
model = SetFitModel.from_pretrained("setfit_model_id")
|
| 183 |
# Run inference
|
| 184 |
-
preds = model("hasCreatedDate: 2024-
|
| 185 |
```
|
| 186 |
|
| 187 |
<!--
|
|
@@ -213,13 +203,13 @@ preds = model("hasCreatedDate: 2024-12-26, hasCustomerHomeCountry: United States
|
|
| 213 |
### Training Set Metrics
|
| 214 |
| Training set | Min | Median | Max |
|
| 215 |
|:-------------|:----|:---------|:----|
|
| 216 |
-
| Word count |
|
| 217 |
|
| 218 |
### Framework Versions
|
| 219 |
- Python: 3.10.16
|
| 220 |
-
- SetFit: 1.1.
|
| 221 |
- Sentence Transformers: 3.4.1
|
| 222 |
-
- Transformers: 4.
|
| 223 |
- PyTorch: 2.6.0+cu124
|
| 224 |
- Datasets: 3.4.1
|
| 225 |
- Tokenizers: 0.21.1
|
|
|
|
| 5 |
- text-classification
|
| 6 |
- generated_from_setfit_trainer
|
| 7 |
widget:
|
| 8 |
+
- text: "hasColourDetails: To be printed on external, adhesive-backed vinyl. No installation\
|
| 9 |
+
\ required. It will be installed from outside the window, so the glue needs to\
|
| 10 |
+
\ be on the reverse., hasCreatedDate: 2024-06-11, hasCustomerHomeCountry: United\
|
| 11 |
+
\ Kingdom, hasCustomerID: 12356, hasCustomerName: Post Office(Post Office), hasCutting:\
|
| 12 |
+
\ Trim to size, hasElementID: 3338891, hasElementTitle: Self-adhesive Window Vinyl\
|
| 13 |
+
\ (no installation), hasFinishedSizeHeight: 150, hasFinishedSizeWidth: 1080, hasInternalID:\
|
| 14 |
+
\ a909113d-a92c-48f2-9817-248a984dc5d8, hasMaterialCategory: Plastic, hasMaterialDescription:\
|
| 15 |
+
\ Suitable self-adhesive Window Vinyl - adhesive on back, hasMaterialType: PVC,\
|
| 16 |
+
\ hasMaterialUnitOfMeasure: GSM, hasNumberOfVersions: 1, hasPackingRequirements:\
|
| 17 |
+
\ Installation is NOT required. Delivery address- FAO Jyoti Rathod, Ayston Road\
|
| 18 |
+
\ Post Office,10 Ayston Road, Leicester LE3 2GA, hasPrice: 30.0, hasPrintedSides:\
|
| 19 |
+
\ Single sided, hasProofType: PDF digital proof, hasQuantity: 1, hasQuantityPerVersion:\
|
| 20 |
+
\ 1, hasSendToDetails: [email protected]., hasSupplierName: Design\
|
| 21 |
+
\ X-Press Limited - CCS Lot 1 Only\t(Design X-Press Limited - CCS Lot 1 Only\t\
|
| 22 |
+
), hasSustainableOptionBeenOffered: N/A, hasTotalColours: 4, hasUnitOfMeasure:\
|
| 23 |
+
\ Millimetres (mm), "
|
| 24 |
+
- text: 'hasAdditionalInformation: 2,600 cards (85 x 55mm), printed one side (full
|
| 25 |
+
colour), delivered to Granby Marketing Services. 26 x packs of 100., hasArtworkDoubleSidedStatus:
|
| 26 |
+
Double Sided Different, hasCreatedDate: 2024-08-22, hasCustomerHomeCountry: United
|
| 27 |
+
Kingdom, hasCustomerID: 29427, hasCustomerName: NHS Blood and Transplant(NHS Blood
|
| 28 |
+
and Transplant), hasCutting: Trim to size, hasElementID: 3466275, hasElementTitle:
|
| 29 |
+
OLC317B Organ Donation Week Cards, hasFinishedSizeHeight: 55, hasFinishedSizeWidth:
|
| 30 |
+
85, hasFscPaperBeenSpecified: No, hasHandFinishing: Yes, hasHandFinishingDetails:
|
| 31 |
+
Shrinkwrap in 100’s, hasInternalID: c171e44e-1c7a-4dba-a6d7-a8c11f317622, hasMaterialCategory:
|
| 32 |
+
Paper, hasMaterialDescription: White Silk Coated Board, hasMaterialRecycledPercentage:
|
| 33 |
+
0%, hasMaterialThicknessOrWeight: 300, hasMaterialType: Paper and board, hasMaterialUnitOfMeasure:
|
| 34 |
+
GSM, hasNumberOfVersions: 1, hasPackingRequirements: DELIVERY, hasPrice: 380.0,
|
| 35 |
+
hasPrintedSides: Double sided, hasProductCategory: Loose Print, hasProofType:
|
| 36 |
+
PDF digital proof, hasQuantity: 2600, hasQuantityPerVersion: 1, hasRecycledContentBeenOffered:
|
| 37 |
+
No, hasSendToDetails: Email to, hasSupplierName: Dataforms Chartered Press Ltd-
|
| 38 |
+
CCS Lot 1 only(Dataforms Chartered Press Ltd- CCS Lot 1 only), hasTotalColours:
|
| 39 |
+
4, hasTotalColoursFace: 4, hasUnitOfMeasure: Millimetres (mm), '
|
| 40 |
+
- text: 'hasCreatedDate: 2024-01-26, hasCustomerHomeCountry: United States, hasCustomerID:
|
| 41 |
+
25570, hasCustomerName: Finish Line Group(Finish Line, Inc), hasCutting: Trim
|
| 42 |
+
to size, hasElementID: 3086661, hasElementTitle: F019753 20 x 28 Fabric Graphic,
|
| 43 |
+
hasFinishedSizeHeight: 28, hasFinishedSizeWidth: 20, hasFscPaperBeenSpecified:
|
| 44 |
+
No, hasInternalID: ac97e6d8-0358-4a1b-9da3-5871bb2b04d1, hasMachineFinishing:
|
| 45 |
+
Yes, hasMachineFinishingDetails: Trim to size. Silicone bed top, bottom left and
|
| 46 |
+
right edges, standard bed, hasMaterialCategory: Textiles, hasMaterialDescription:
|
| 47 |
+
Moonlight Itex - 260GSM, hasMaterialRecycledPercentage: 0%, hasMaterialType: Polyester,
|
| 48 |
+
hasNumberOfVersions: 1, hasPackingRequirements: Trim to size. Silicone bed top,
|
| 49 |
+
bottom left and right edges, standard bed, hasPrice: 1395.56, hasPrintedSides:
|
| 50 |
+
Single sided, hasProductCategory: Displays - Backlit, hasProofType: PDF digital
|
| 51 |
+
proof, hasQuantity: 139, hasRecycledContentBeenOffered: N/A, hasSupplierName:
|
| 52 |
+
GSP Custom Color(GSP Custom Color - HHGSP), hasTotalColours: 4, hasUnitOfMeasure:
|
| 53 |
+
Inches (in), '
|
| 54 |
+
- text: 'hasAdditionalInformation: AO, hasCreatedDate: 2024-10-08, hasCustomerHomeCountry:
|
| 55 |
+
United States, hasCustomerID: 30642, hasCustomerName: Station Casinos LLC(Station
|
| 56 |
+
Casinos), hasCutting: Trim to size, hasElementID: 3555960, hasElementTitle: 211696
|
| 57 |
+
- 381" X 363" SS FRONTLIT EXTERIOR VINYL SIGN , hasFinishedSizeHeight: 363, hasFinishedSizeWidth:
|
| 58 |
+
381, hasFscPaperBeenSpecified: No, hasInternalID: 04b8890a-dc33-4778-ad73-c1f68f68231c,
|
| 59 |
+
hasMaterialCategory: Plastic, hasMaterialDescription: 13OZ VINYL, hasMaterialRecycledPercentage:
|
| 60 |
+
0%, hasMaterialThicknessOrWeight: 13, hasMaterialType: PVC, hasMaterialUnitOfMeasure:
|
| 61 |
+
Ounces (oz), hasNumberOfVersions: 1, hasPrice: 676.0, hasPrintedSides: Not printed,
|
| 62 |
+
hasProductCategory: Banners (synthetic), hasProofType: PDF digital proof, hasQuantity:
|
| 63 |
+
1, hasRecycledContentBeenOffered: N/A, hasSupplierName: WestRock Company(Westrock
|
| 64 |
+
- 14360 - HHGSP), hasTotalColours: 4, hasUnitOfMeasure: Inches (in), '
|
| 65 |
+
- text: 'hasAdditionalInformation: 2 - Pages, hasArtworkDoubleSidedStatus: Double
|
| 66 |
+
Sided Different, hasColourDetails: 1/1 (Black Two-sided), hasCreatedDate: 2024-12-10,
|
| 67 |
+
hasCustomerHomeCountry: United States, hasCustomerID: 26760, hasCustomerName:
|
| 68 |
+
Elanco Animal Health(Elanco Animal Health), hasCutting: Trim to size, hasElementID:
|
| 69 |
+
3687940, hasElementTitle: PA103754X - Galliprant PI, hasFinishedSizeHeight: 11,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 70 |
hasFinishedSizeWidth: 8.5, hasFlatSizeHeight: 11, hasFlatSizeWidth: 8.5, hasFscPaperBeenSpecified:
|
| 71 |
+
No, hasInternalID: c0585f9a-6716-4373-a218-041b92baf4a2, hasMachineFinishing:
|
| 72 |
+
Yes, hasMachineFinishingDetails: bleed, hasMaterialCategory: Paper, hasMaterialDescription:
|
| 73 |
+
60# White Offset, hasMaterialThicknessOrWeight: 60, hasMaterialType: Paper and
|
| 74 |
+
board, hasMaterialUnitOfMeasure: Pounds (lbs), hasMinimumRecycledContent: 0%,
|
| 75 |
+
hasNumberOfVersions: 1, hasPackingRequirements: Hold for Kit Packing in Element
|
| 76 |
+
5, hasPaperType: Offset, hasPrice: 350.0, hasPrintedSides: Double sided, hasProductCategory:
|
| 77 |
+
Booklets & Brochures, hasProofType: PDF digital proof, hasQuantity: 3100, hasRecycledContentBeenRequested:
|
| 78 |
+
No, hasSupplierName: Modern Litho – Kansas City(Modernlitho -James Printing.
|
| 79 |
+
Inc - HHGSP), hasTotalColoursFace: 1, hasTotalColoursReverse: 1, hasUnitOfMeasure:
|
| 80 |
+
Inches (in), '
|
|
|
|
|
|
|
| 81 |
metrics:
|
| 82 |
- f1_micro
|
| 83 |
- f1_macro
|
|
|
|
| 100 |
split: test
|
| 101 |
metrics:
|
| 102 |
- type: f1_micro
|
| 103 |
+
value: 0.9322709163346613
|
| 104 |
name: F1_Micro
|
| 105 |
- type: f1_macro
|
| 106 |
+
value: 0.37012987012987014
|
| 107 |
name: F1_Macro
|
| 108 |
- type: f1_weighted
|
| 109 |
+
value: 0.8821255080797066
|
| 110 |
name: F1_Weighted
|
| 111 |
- type: precision
|
| 112 |
+
value: 0.9750000238418579
|
| 113 |
name: Precision
|
| 114 |
- type: accuracy
|
| 115 |
+
value: 0.9468749761581421
|
| 116 |
name: Accuracy
|
| 117 |
- type: recall
|
| 118 |
+
value: 0.8931297659873962
|
| 119 |
name: Recall
|
| 120 |
---
|
| 121 |
|
|
|
|
| 151 |
### Metrics
|
| 152 |
| Label | F1_Micro | F1_Macro | F1_Weighted | Precision | Accuracy | Recall |
|
| 153 |
|:--------|:---------|:---------|:------------|:----------|:---------|:-------|
|
| 154 |
+
| **all** | 0.9323 | 0.3701 | 0.8821 | 0.9750 | 0.9469 | 0.8931 |
|
| 155 |
|
| 156 |
## Uses
|
| 157 |
|
|
|
|
| 171 |
# Download from the 🤗 Hub
|
| 172 |
model = SetFitModel.from_pretrained("setfit_model_id")
|
| 173 |
# Run inference
|
| 174 |
+
preds = model("hasAdditionalInformation: AO, hasCreatedDate: 2024-10-08, hasCustomerHomeCountry: United States, hasCustomerID: 30642, hasCustomerName: Station Casinos LLC(Station Casinos), hasCutting: Trim to size, hasElementID: 3555960, hasElementTitle: 211696 - 381\" X 363\" SS FRONTLIT EXTERIOR VINYL SIGN , hasFinishedSizeHeight: 363, hasFinishedSizeWidth: 381, hasFscPaperBeenSpecified: No, hasInternalID: 04b8890a-dc33-4778-ad73-c1f68f68231c, hasMaterialCategory: Plastic, hasMaterialDescription: 13OZ VINYL, hasMaterialRecycledPercentage: 0%, hasMaterialThicknessOrWeight: 13, hasMaterialType: PVC, hasMaterialUnitOfMeasure: Ounces (oz), hasNumberOfVersions: 1, hasPrice: 676.0, hasPrintedSides: Not printed, hasProductCategory: Banners (synthetic), hasProofType: PDF digital proof, hasQuantity: 1, hasRecycledContentBeenOffered: N/A, hasSupplierName: WestRock Company(Westrock - 14360 - HHGSP), hasTotalColours: 4, hasUnitOfMeasure: Inches (in), ")
|
| 175 |
```
|
| 176 |
|
| 177 |
<!--
|
|
|
|
| 203 |
### Training Set Metrics
|
| 204 |
| Training set | Min | Median | Max |
|
| 205 |
|:-------------|:----|:---------|:----|
|
| 206 |
+
| Word count | 67 | 110.9875 | 238 |
|
| 207 |
|
| 208 |
### Framework Versions
|
| 209 |
- Python: 3.10.16
|
| 210 |
+
- SetFit: 1.1.2
|
| 211 |
- Sentence Transformers: 3.4.1
|
| 212 |
+
- Transformers: 4.51.3
|
| 213 |
- PyTorch: 2.6.0+cu124
|
| 214 |
- Datasets: 3.4.1
|
| 215 |
- Tokenizers: 0.21.1
|
config.json
CHANGED
|
@@ -23,7 +23,7 @@
|
|
| 23 |
"pad_token_id": 0,
|
| 24 |
"position_embedding_type": "absolute",
|
| 25 |
"torch_dtype": "float32",
|
| 26 |
-
"transformers_version": "4.
|
| 27 |
"type_vocab_size": 2,
|
| 28 |
"use_cache": true,
|
| 29 |
"vocab_size": 30522
|
|
|
|
| 23 |
"pad_token_id": 0,
|
| 24 |
"position_embedding_type": "absolute",
|
| 25 |
"torch_dtype": "float32",
|
| 26 |
+
"transformers_version": "4.51.3",
|
| 27 |
"type_vocab_size": 2,
|
| 28 |
"use_cache": true,
|
| 29 |
"vocab_size": 30522
|
config_sentence_transformers.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
"__version__": {
|
| 3 |
"sentence_transformers": "3.4.1",
|
| 4 |
-
"transformers": "4.
|
| 5 |
"pytorch": "2.6.0+cu124"
|
| 6 |
},
|
| 7 |
"prompts": {},
|
|
|
|
| 1 |
{
|
| 2 |
"__version__": {
|
| 3 |
"sentence_transformers": "3.4.1",
|
| 4 |
+
"transformers": "4.51.3",
|
| 5 |
"pytorch": "2.6.0+cu124"
|
| 6 |
},
|
| 7 |
"prompts": {},
|
config_setfit.json
CHANGED
|
@@ -1,12 +1,12 @@
|
|
| 1 |
{
|
| 2 |
"labels": [
|
| 3 |
"SupplierDomain",
|
|
|
|
| 4 |
"ManufacturingDomain",
|
|
|
|
|
|
|
| 5 |
"FeatureDomain",
|
| 6 |
"ProductDomain",
|
| 7 |
-
"MaterialDomain",
|
| 8 |
-
"AssemblyDomain",
|
| 9 |
-
"MeasurementDomain",
|
| 10 |
"ShippingAndHandlingDomain"
|
| 11 |
],
|
| 12 |
"normalize_embeddings": false
|
|
|
|
| 1 |
{
|
| 2 |
"labels": [
|
| 3 |
"SupplierDomain",
|
| 4 |
+
"AssemblyDomain",
|
| 5 |
"ManufacturingDomain",
|
| 6 |
+
"MeasurementDomain",
|
| 7 |
+
"MaterialDomain",
|
| 8 |
"FeatureDomain",
|
| 9 |
"ProductDomain",
|
|
|
|
|
|
|
|
|
|
| 10 |
"ShippingAndHandlingDomain"
|
| 11 |
],
|
| 12 |
"normalize_embeddings": false
|
heads/domain-router.pkl
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:95bcfbf81f9a6390af8bacf1be63d10e8c597eaf70b62fcd241b979ae4cd8c72
|
| 3 |
+
size 29664
|
heads/materials-classifier.pkl
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:12f601cd607530a892a44d5c83faa1e9e6fafe755bafb3a6958a1277bf8cef3b
|
| 3 |
+
size 86247
|
model_head.pkl
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:371d6e5daf5532c6e7da4474e59efb000863718ce8465b7e2e1ae6d914765940
|
| 3 |
+
size 28404
|