Hugging Face's logo

weizhou03
/
nomic-embed-text-v1.5

Text Classification
setfit
Safetensors
GGUF
sentence-transformers
nomic_bert
generated_from_setfit_trainer
custom_code
Eval Results
feature-extraction
Model card Files
xet
Community
weizhou03 commited on
Commit
f5f1c05
·
verified ·
1 Parent(s): 7203cd0

Push model using huggingface_hub.

Browse files
Files changed (3) hide show
  1. README.md +416 -146
  2. model.safetensors +1 -1
  3. model_head.pkl +1 -1
README.md CHANGED
@@ -5,11 +5,11 @@ tags:
5
  - text-classification
6
  - generated_from_setfit_trainer
7
  widget:
8
- - text: I need some icon suggestions for this layout
9
- - text: Tighten the letter spacing
10
- - text: Group the logo and title together
 
11
  - text: Create a photo of a mountain landscape
12
- - text: Mirror the logo vertically
13
  metrics:
14
  - accuracy
15
  pipeline_tag: text-classification
@@ -28,7 +28,7 @@ model-index:
28
  split: test
29
  metrics:
30
  - type: accuracy
31
- value: 0.2895622895622896
32
  name: Accuracy
33
  ---
34
 
@@ -60,78 +60,78 @@ The model has been trained using an efficient few-shot learning technique that i
60
  - **Blogpost:** [SetFit: Efficient Few-Shot Learning Without Prompts](https://huggingface.co/blog/setfit)
61
 
62
  ### Model Labels
63
- | Label | Examples |
64
- |:------|:-------------------------------------------------------------------|
65
- | 0 | <ul><li>'Add a corporate presentation background'</li></ul> |
66
- | 1 | <ul><li>'I need some icon suggestions for this layout'</li></ul> |
67
- | 2 | <ul><li>"Add a heading that says 'Welcome'"</li></ul> |
68
- | 3 | <ul><li>'Distribute the shapes across the page'</li></ul> |
69
- | 4 | <ul><li>'Add a zoom animation to the logo'</li></ul> |
70
- | 5 | <ul><li>'Make everything fade in gradually'</li></ul> |
71
- | 6 | <ul><li>'Remove the unwanted text overlay'</li></ul> |
72
- | 7 | <ul><li>'Remove the background shape'</li></ul> |
73
- | 8 | <ul><li>'What shape options do you have?'</li></ul> |
74
- | 9 | <ul><li>'Distribute the buttons around the center image'</li></ul> |
75
- | 10 | <ul><li>'Duplicate the page structure'</li></ul> |
76
- | 11 | <ul><li>'Duplicate the icon and move it'</li></ul> |
77
- | 12 | <ul><li>'Copy the content to the final page'</li></ul> |
78
- | 13 | <ul><li>'Fix the letter spacing'</li></ul> |
79
- | 14 | <ul><li>'Mirror the logo vertically'</li></ul> |
80
- | 15 | <ul><li>'Create a photo of a mountain landscape'</li></ul> |
81
- | 16 | <ul><li>'Create a card for a birthday party'</li></ul> |
82
- | 17 | <ul><li>'Group the logo and title together'</li></ul> |
83
- | 18 | <ul><li>'Move the image to the center'</li></ul> |
84
- | 19 | <ul><li>'Apply a vintage filter to the photo'</li></ul> |
85
- | 20 | <ul><li>'Find me texture patterns'</li></ul> |
86
- | 21 | <ul><li>'Restore the previous color'</li></ul> |
87
- | 22 | <ul><li>'Remove the background from the illustration'</li></ul> |
88
- | 23 | <ul><li>'Delete the old sign'</li></ul> |
89
- | 24 | <ul><li>'Replace the old logo with a new one'</li></ul> |
90
- | 25 | <ul><li>'Replace the tagline'</li></ul> |
91
- | 26 | <ul><li>'Reset the image adjustments'</li></ul> |
92
- | 27 | <ul><li>'Scale the text up'</li></ul> |
93
- | 28 | <ul><li>'Resize to LinkedIn post dimensions'</li></ul> |
94
- | 29 | <ul><li>'Turn the logo 270 degrees'</li></ul> |
95
- | 30 | <ul><li>'Scatter the particles around the text'</li></ul> |
96
- | 31 | <ul><li>'Select the footer content'</li></ul> |
97
- | 32 | <ul><li>'Change to a light gray background'</li></ul> |
98
- | 33 | <ul><li>'Change the blend mode to color burn'</li></ul> |
99
- | 34 | <ul><li>'Blur the logo slightly'</li></ul> |
100
- | 35 | <ul><li>'Add a double border to the image'</li></ul> |
101
- | 36 | <ul><li>'Make the photo more illuminated'</li></ul> |
102
- | 37 | <ul><li>'Send the logo to the back'</li></ul> |
103
- | 38 | <ul><li>'Increase the image contrast'</li></ul> |
104
- | 39 | <ul><li>'Crop to a polygon'</li></ul> |
105
- | 40 | <ul><li>'Add a gradient shadow'</li></ul> |
106
- | 41 | <ul><li>'Change the background color to green'</li></ul> |
107
- | 42 | <ul><li>'Increase the menu font size'</li></ul> |
108
- | 43 | <ul><li>'Make the text bold and strikethrough'</li></ul> |
109
- | 44 | <ul><li>'Use a decorative font for the logo'</li></ul> |
110
- | 45 | <ul><li>'Brighten the light areas'</li></ul> |
111
- | 46 | <ul><li>'Set the picture as full background'</li></ul> |
112
- | 47 | <ul><li>'Tighten the letter spacing'</li></ul> |
113
- | 48 | <ul><li>'Add more line height'</li></ul> |
114
- | 49 | <ul><li>'Reduce the opacity of the overlay'</li></ul> |
115
- | 50 | <ul><li>'Reduce the paragraph spacing'</li></ul> |
116
- | 51 | <ul><li>'Make the colors more vibrant'</li></ul> |
117
- | 52 | <ul><li>'Make the shadows deeper'</li></ul> |
118
- | 53 | <ul><li>'Increase the image sharpness'</li></ul> |
119
- | 54 | <ul><li>'Left align the subtitle'</li></ul> |
120
- | 55 | <ul><li>'Create a text container'</li></ul> |
121
- | 56 | <ul><li>'Create text in a spiral pattern'</li></ul> |
122
- | 57 | <ul><li>'Convert to bullet point format'</li></ul> |
123
- | 58 | <ul><li>'Add a colored shadow to the text'</li></ul> |
124
- | 59 | <ul><li>'Increase the warm color cast'</li></ul> |
125
- | 60 | <ul><li>'I want to upload a new image'</li></ul> |
126
- | 61 | <ul><li>'Revert the opacity change'</li></ul> |
127
- | 62 | <ul><li>'Break apart the grouped elements'</li></ul> |
128
 
129
  ## Evaluation
130
 
131
  ### Metrics
132
  | Label | Accuracy |
133
  |:--------|:---------|
134
- | **all** | 0.2896 |
135
 
136
  ## Uses
137
 
@@ -151,7 +151,7 @@ from setfit import SetFitModel
151
  # Download from the 🤗 Hub
152
  model = SetFitModel.from_pretrained("setfit_model_id")
153
  # Run inference
154
- preds = model("Tighten the letter spacing")
155
  ```
156
 
157
  <!--
@@ -183,77 +183,77 @@ preds = model("Tighten the letter spacing")
183
  ### Training Set Metrics
184
  | Training set | Min | Median | Max |
185
  |:-------------|:----|:-------|:----|
186
- | Word count | 3 | 5.2857 | 8 |
187
 
188
  | Label | Training Sample Count |
189
  |:------|:----------------------|
190
- | 0 | 1 |
191
- | 1 | 1 |
192
- | 2 | 1 |
193
- | 3 | 1 |
194
- | 4 | 1 |
195
- | 5 | 1 |
196
- | 6 | 1 |
197
- | 7 | 1 |
198
- | 8 | 1 |
199
- | 9 | 1 |
200
- | 10 | 1 |
201
- | 11 | 1 |
202
- | 12 | 1 |
203
- | 13 | 1 |
204
- | 14 | 1 |
205
- | 15 | 1 |
206
- | 16 | 1 |
207
- | 17 | 1 |
208
- | 18 | 1 |
209
- | 19 | 1 |
210
- | 20 | 1 |
211
- | 21 | 1 |
212
- | 22 | 1 |
213
- | 23 | 1 |
214
- | 24 | 1 |
215
- | 25 | 1 |
216
- | 26 | 1 |
217
- | 27 | 1 |
218
- | 28 | 1 |
219
- | 29 | 1 |
220
- | 30 | 1 |
221
- | 31 | 1 |
222
- | 32 | 1 |
223
- | 33 | 1 |
224
- | 34 | 1 |
225
- | 35 | 1 |
226
- | 36 | 1 |
227
- | 37 | 1 |
228
- | 38 | 1 |
229
- | 39 | 1 |
230
- | 40 | 1 |
231
- | 41 | 1 |
232
- | 42 | 1 |
233
- | 43 | 1 |
234
- | 44 | 1 |
235
- | 45 | 1 |
236
- | 46 | 1 |
237
- | 47 | 1 |
238
- | 48 | 1 |
239
- | 49 | 1 |
240
- | 50 | 1 |
241
- | 51 | 1 |
242
- | 52 | 1 |
243
- | 53 | 1 |
244
- | 54 | 1 |
245
- | 55 | 1 |
246
- | 56 | 1 |
247
- | 57 | 1 |
248
- | 58 | 1 |
249
- | 59 | 1 |
250
- | 60 | 1 |
251
- | 61 | 1 |
252
- | 62 | 1 |
253
 
254
  ### Training Hyperparameters
255
- - batch_size: (32, 32)
256
- - num_epochs: (2, 2)
257
  - max_steps: -1
258
  - sampling_strategy: oversampling
259
  - body_learning_rate: (2e-05, 1e-05)
@@ -261,7 +261,7 @@ preds = model("Tighten the letter spacing")
261
  - loss: CosineSimilarityLoss
262
  - distance_metric: cosine_distance
263
  - margin: 0.25
264
- - end_to_end: False
265
  - use_amp: False
266
  - warmup_proportion: 0.1
267
  - l2_weight: 0.1
@@ -270,13 +270,283 @@ preds = model("Tighten the letter spacing")
270
  - load_best_model_at_end: False
271
 
272
  ### Training Results
273
- | Epoch | Step | Training Loss | Validation Loss |
274
- |:------:|:----:|:-------------:|:---------------:|
275
- | 0.0081 | 1 | 0.1179 | - |
276
- | 0.4065 | 50 | 0.0211 | - |
277
- | 0.8130 | 100 | 0.002 | - |
278
- | 1.2195 | 150 | 0.0011 | - |
279
- | 1.6260 | 200 | 0.0011 | - |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
280
 
281
  ### Framework Versions
282
  - Python: 3.12.11
 
5
  - text-classification
6
  - generated_from_setfit_trainer
7
  widget:
8
+ - text: Show me some texture backgrounds
9
+ - text: Add more space between paragraphs
10
+ - text: Position the text at the top
11
+ - text: Create an Instagram story for a birthday
12
  - text: Create a photo of a mountain landscape
 
13
  metrics:
14
  - accuracy
15
  pipeline_tag: text-classification
 
28
  split: test
29
  metrics:
30
  - type: accuracy
31
+ value: 0.6240179573512907
32
  name: Accuracy
33
  ---
34
 
 
60
  - **Blogpost:** [SetFit: Efficient Few-Shot Learning Without Prompts](https://huggingface.co/blog/setfit)
61
 
62
  ### Model Labels
63
+ | Label | Examples |
64
+ |:------|:------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
65
+ | 0 | <ul><li>'Add a corporate presentation background'</li><li>'Insert a modern icon set for the design'</li><li>'Add a mountain landscape background to the page'</li></ul> |
66
+ | 1 | <ul><li>'Show me some background options for my design'</li><li>'I need some icon suggestions for this layout'</li><li>'Find me some shape options for this design'</li></ul> |
67
+ | 2 | <ul><li>'Add a social media handle'</li><li>'Insert a promotional message'</li><li>"Add a heading that says 'Welcome'"</li></ul> |
68
+ | 3 | <ul><li>'Align the footer elements'</li><li>'Align everything to the bottom'</li><li>'Center the logo and title'</li></ul> |
69
+ | 4 | <ul><li>'Make the image slide from left'</li><li>'Make the image bounce in'</li><li>'Add a fade-in animation to the text'</li></ul> |
70
+ | 5 | <ul><li>'Make the page rotate in'</li><li>'Add a zoom-out effect to the entire page'</li><li>'Make everything fade in gradually'</li></ul> |
71
+ | 6 | <ul><li>'Change the color of the car to red'</li><li>'Remove the unwanted text overlay'</li><li>'Remove the sunglasses and add regular glasses'</li></ul> |
72
+ | 7 | <ul><li>'Remove the unnecessary decoration'</li><li>'Remove the temporary shape'</li><li>'Delete the broken image'</li></ul> |
73
+ | 8 | <ul><li>'How do I group elements together?'</li><li>'What text formatting options are there?'</li><li>'What are the available filter effects?'</li></ul> |
74
+ | 9 | <ul><li>'Place the decorations around the title'</li><li>'Distribute the text around the focal point'</li><li>'Arrange the icons in a circle around the logo'</li></ul> |
75
+ | 10 | <ul><li>'Make a copy for editing'</li><li>'Create a similar page'</li><li>'Make a backup of this page'</li></ul> |
76
+ | 11 | <ul><li>'Copy the image and make it smaller'</li><li>'Duplicate the text and change the color'</li><li>'Duplicate the logo and flip it'</li></ul> |
77
+ | 12 | <ul><li>'Copy the icon to the next slide'</li><li>'Duplicate the button to page 5'</li><li>'Copy the text to the last page'</li></ul> |
78
+ | 13 | <ul><li>'Fix the typographic errors'</li><li>'Fix the letter spacing'</li><li>'Improve the paragraph spacing'</li></ul> |
79
+ | 14 | <ul><li>'Mirror the logo vertically'</li><li>'Mirror the icon horizontally'</li><li>'Flip the image right to left'</li></ul> |
80
+ | 15 | <ul><li>'Make an image of a cozy coffee shop'</li><li>'Make an image of a modern office space'</li><li>'Make an image of a magical forest'</li></ul> |
81
+ | 16 | <ul><li>'Create a card for a birthday party'</li><li>'Create a dark-themed social media post for tech news'</li><li>'Generate an Instagram post for a birthday'</li></ul> |
82
+ | 17 | <ul><li>'Group the navigation elements'</li><li>'Group the header elements'</li><li>'Group the decorative elements'</li></ul> |
83
+ | 18 | <ul><li>'Move the element to the bottom left'</li><li>'Move the text to the middle'</li><li>'Position the logo in the corner'</li></ul> |
84
+ | 19 | <ul><li>'Apply a cool blue filter'</li><li>'Add a dreamy effect'</li><li>'Apply a cinematic filter'</li></ul> |
85
+ | 20 | <ul><li>'Find me some background images'</li><li>'Suggest some visual elements'</li><li>'Find me overlay effects'</li></ul> |
86
+ | 21 | <ul><li>'Redo the filter effect'</li><li>'Bring back the deleted element'</li><li>'Bring back the deleted shape'</li></ul> |
87
+ | 22 | <ul><li>'Make the object stand alone'</li><li>'Remove the background from the icon'</li><li>'Remove the background from the photo'</li></ul> |
88
+ | 23 | <ul><li>'Remove the distracting element'</li><li>'Delete the watermark'</li><li>'Remove the old building'</li></ul> |
89
+ | 24 | <ul><li>'Replace the image with a better one'</li><li>'Change the background to a different scene'</li><li>'Replace the overlay with a gradient'</li></ul> |
90
+ | 25 | <ul><li>'Replace the menu items'</li><li>'Replace the title'</li><li>'Change the button text'</li></ul> |
91
+ | 26 | <ul><li>'Restore the original brightness'</li><li>'Reset the image to its initial state'</li><li>'Restore the original colors'</li></ul> |
92
+ | 27 | <ul><li>'Reduce the shape size'</li><li>'Increase the image size'</li><li>'Reduce the button size'</li></ul> |
93
+ | 28 | <ul><li>'Make the page portrait orientation'</li><li>'Change to business card size'</li><li>'Make the page wider'</li></ul> |
94
+ | 29 | <ul><li>'Rotate the element 60 degrees'</li><li>'Rotate the image clockwise'</li><li>'Turn the logo 270 degrees'</li></ul> |
95
+ | 30 | <ul><li>'Scatter the flowers around the border'</li><li>'Distribute the shapes randomly'</li><li>'Scatter the stars around the title'</li></ul> |
96
+ | 31 | <ul><li>'Select the main image'</li><li>'Choose the heading text'</li><li>'Select the logo element'</li></ul> |
97
+ | 32 | <ul><li>'Make the background transparent'</li><li>'Change to a neutral background'</li><li>'Change to a light gray background'</li></ul> |
98
+ | 33 | <ul><li>'Change the blend mode to exclusion'</li><li>'Set the blend mode to soft light'</li><li>'Set the blend mode to darken'</li></ul> |
99
+ | 34 | <ul><li>'Add a depth of field blur'</li><li>'Blur the entire image'</li><li>'Add a motion blur effect'</li></ul> |
100
+ | 35 | <ul><li>'Change the border width'</li><li>'Add a colored border'</li><li>'Add a gradient border'</li></ul> |
101
+ | 36 | <ul><li>'Brighten the dark areas'</li><li>'Make the photo more illuminated'</li><li>'Make the image more brilliant'</li></ul> |
102
+ | 37 | <ul><li>'Put the decoration behind'</li><li>'Move the icon to the top'</li><li>'Put the shape behind the text'</li></ul> |
103
+ | 38 | <ul><li>'Make the image more striking'</li><li>'Enhance the contrast'</li><li>'Increase the visual clarity'</li></ul> |
104
+ | 39 | <ul><li>'Make the image elliptical'</li><li>'Make the image circular'</li><li>'Make the image diamond-shaped'</li></ul> |
105
+ | 40 | <ul><li>'Add a colored drop shadow'</li><li>'Create a realistic shadow'</li><li>'Create a soft drop shadow'</li></ul> |
106
+ | 41 | <ul><li>'Fill the shape with orange'</li><li>'Change the color to teal'</li><li>'Fill the element with brown'</li></ul> |
107
+ | 42 | <ul><li>'Make the subtitle larger'</li><li>'Make the description smaller'</li><li>'Make the label smaller'</li></ul> |
108
+ | 43 | <ul><li>'Make the text italic only'</li><li>'Make the text emphasized'</li><li>'Make the text bold only'</li></ul> |
109
+ | 44 | <ul><li>'Use a playful font'</li><li>'Use a professional font'</li><li>'Change to a modern sans-serif font'</li></ul> |
110
+ | 45 | <ul><li>'Increase the highlight clarity'</li><li>'Increase the highlight intensity'</li><li>'Enhance the bright spots'</li></ul> |
111
+ | 46 | <ul><li>'Make the photo cover the background'</li><li>'Set the picture as full background'</li><li>'Make the photo cover the entire page'</li></ul> |
112
+ | 47 | <ul><li>'Spread out the letters in the heading'</li><li>'Make the letters closer together in the text'</li><li>'Tighten the letter spacing in the caption'</li></ul> |
113
+ | 48 | <ul><li>'Add more line height'</li><li>'Make the text more readable'</li><li>'Reduce the line height'</li></ul> |
114
+ | 49 | <ul><li>'Make the text more transparent'</li><li>'Reduce the opacity of the overlay'</li><li>'Make the element semi-transparent'</li></ul> |
115
+ | 50 | <ul><li>'Make the sections closer'</li><li>'Make the paragraphs closer together'</li><li>'Reduce the paragraph spacing'</li></ul> |
116
+ | 51 | <ul><li>'Make the colors more vivid'</li><li>'Increase the color intensity'</li><li>'Increase the color richness'</li></ul> |
117
+ | 52 | <ul><li>'Darken the shadow regions'</li><li>'Darken the shadow tones'</li><li>'Enhance the shadow details'</li></ul> |
118
+ | 53 | <ul><li>'Make the details more sharp'</li><li>'Increase the image resolution'</li><li>'Increase the detail sharpness'</li></ul> |
119
+ | 54 | <ul><li>'Center the title text'</li><li>'Center the logo text'</li><li>'Justify the paragraph text'</li></ul> |
120
+ | 55 | <ul><li>'Create a text container'</li><li>'Create a border around the text'</li><li>'Add a gradient background to the text'</li></ul> |
121
+ | 56 | <ul><li>'Create curved text layout'</li><li>'Make the text flow in an arc'</li><li>'Make the text follow a path'</li></ul> |
122
+ | 57 | <ul><li>'Make the text into bullet points'</li><li>'Make the text into a list with bullets'</li><li>'Create an organized list from the text'</li></ul> |
123
+ | 58 | <ul><li>'Create a shadow behind the heading'</li><li>'Add a dramatic text shadow'</li><li>'Add a soft glow effect'</li></ul> |
124
+ | 59 | <ul><li>'Increase the warm color cast'</li><li>'Make the image more yellow'</li><li>'Add warm undertones to the photo'</li></ul> |
125
+ | 60 | <ul><li>'Open the image upload tool'</li><li>'Show me the file browser'</li><li>'Open the upload window'</li></ul> |
126
+ | 61 | <ul><li>'Go back to the previous state'</li><li>'Undo the text edit'</li><li>'Revert the color change'</li></ul> |
127
+ | 62 | <ul><li>'Ungroup the selected items'</li><li>'Break up the grouped objects'</li><li>'Ungroup the merged elements'</li></ul> |
128
 
129
  ## Evaluation
130
 
131
  ### Metrics
132
  | Label | Accuracy |
133
  |:--------|:---------|
134
+ | **all** | 0.6240 |
135
 
136
  ## Uses
137
 
 
151
  # Download from the 🤗 Hub
152
  model = SetFitModel.from_pretrained("setfit_model_id")
153
  # Run inference
154
+ preds = model("Position the text at the top")
155
  ```
156
 
157
  <!--
 
183
  ### Training Set Metrics
184
  | Training set | Min | Median | Max |
185
  |:-------------|:----|:-------|:----|
186
+ | Word count | 3 | 5.2243 | 12 |
187
 
188
  | Label | Training Sample Count |
189
  |:------|:----------------------|
190
+ | 0 | 15 |
191
+ | 1 | 15 |
192
+ | 2 | 15 |
193
+ | 3 | 15 |
194
+ | 4 | 15 |
195
+ | 5 | 15 |
196
+ | 6 | 15 |
197
+ | 7 | 15 |
198
+ | 8 | 15 |
199
+ | 9 | 15 |
200
+ | 10 | 15 |
201
+ | 11 | 15 |
202
+ | 12 | 15 |
203
+ | 13 | 15 |
204
+ | 14 | 15 |
205
+ | 15 | 15 |
206
+ | 16 | 15 |
207
+ | 17 | 15 |
208
+ | 18 | 15 |
209
+ | 19 | 15 |
210
+ | 20 | 15 |
211
+ | 21 | 15 |
212
+ | 22 | 15 |
213
+ | 23 | 15 |
214
+ | 24 | 15 |
215
+ | 25 | 15 |
216
+ | 26 | 15 |
217
+ | 27 | 15 |
218
+ | 28 | 15 |
219
+ | 29 | 15 |
220
+ | 30 | 15 |
221
+ | 31 | 15 |
222
+ | 32 | 15 |
223
+ | 33 | 15 |
224
+ | 34 | 15 |
225
+ | 35 | 15 |
226
+ | 36 | 15 |
227
+ | 37 | 15 |
228
+ | 38 | 15 |
229
+ | 39 | 15 |
230
+ | 40 | 15 |
231
+ | 41 | 15 |
232
+ | 42 | 15 |
233
+ | 43 | 15 |
234
+ | 44 | 15 |
235
+ | 45 | 15 |
236
+ | 46 | 15 |
237
+ | 47 | 15 |
238
+ | 48 | 15 |
239
+ | 49 | 15 |
240
+ | 50 | 15 |
241
+ | 51 | 15 |
242
+ | 52 | 15 |
243
+ | 53 | 15 |
244
+ | 54 | 15 |
245
+ | 55 | 15 |
246
+ | 56 | 15 |
247
+ | 57 | 15 |
248
+ | 58 | 15 |
249
+ | 59 | 15 |
250
+ | 60 | 15 |
251
+ | 61 | 15 |
252
+ | 62 | 15 |
253
 
254
  ### Training Hyperparameters
255
+ - batch_size: (64, 64)
256
+ - num_epochs: (1, 1)
257
  - max_steps: -1
258
  - sampling_strategy: oversampling
259
  - body_learning_rate: (2e-05, 1e-05)
 
261
  - loss: CosineSimilarityLoss
262
  - distance_metric: cosine_distance
263
  - margin: 0.25
264
+ - end_to_end: True
265
  - use_amp: False
266
  - warmup_proportion: 0.1
267
  - l2_weight: 0.1
 
270
  - load_best_model_at_end: False
271
 
272
  ### Training Results
273
+ | Epoch | Step | Training Loss | Validation Loss |
274
+ |:------:|:-----:|:-------------:|:---------------:|
275
+ | 0.0001 | 1 | 0.1774 | - |
276
+ | 0.0036 | 50 | 0.1629 | - |
277
+ | 0.0073 | 100 | 0.1464 | - |
278
+ | 0.0109 | 150 | 0.1147 | - |
279
+ | 0.0146 | 200 | 0.0798 | - |
280
+ | 0.0182 | 250 | 0.0552 | - |
281
+ | 0.0218 | 300 | 0.0391 | - |
282
+ | 0.0255 | 350 | 0.0271 | - |
283
+ | 0.0291 | 400 | 0.0272 | - |
284
+ | 0.0328 | 450 | 0.018 | - |
285
+ | 0.0364 | 500 | 0.015 | - |
286
+ | 0.0400 | 550 | 0.0136 | - |
287
+ | 0.0437 | 600 | 0.012 | - |
288
+ | 0.0473 | 650 | 0.0105 | - |
289
+ | 0.0510 | 700 | 0.0094 | - |
290
+ | 0.0546 | 750 | 0.0087 | - |
291
+ | 0.0583 | 800 | 0.0061 | - |
292
+ | 0.0619 | 850 | 0.0078 | - |
293
+ | 0.0655 | 900 | 0.0068 | - |
294
+ | 0.0692 | 950 | 0.0066 | - |
295
+ | 0.0728 | 1000 | 0.0053 | - |
296
+ | 0.0765 | 1050 | 0.0057 | - |
297
+ | 0.0801 | 1100 | 0.0065 | - |
298
+ | 0.0837 | 1150 | 0.0054 | - |
299
+ | 0.0874 | 1200 | 0.0058 | - |
300
+ | 0.0910 | 1250 | 0.006 | - |
301
+ | 0.0947 | 1300 | 0.0048 | - |
302
+ | 0.0983 | 1350 | 0.0038 | - |
303
+ | 0.1019 | 1400 | 0.0034 | - |
304
+ | 0.1056 | 1450 | 0.0037 | - |
305
+ | 0.1092 | 1500 | 0.006 | - |
306
+ | 0.1129 | 1550 | 0.0047 | - |
307
+ | 0.1165 | 1600 | 0.0042 | - |
308
+ | 0.1201 | 1650 | 0.0038 | - |
309
+ | 0.1238 | 1700 | 0.0036 | - |
310
+ | 0.1274 | 1750 | 0.0049 | - |
311
+ | 0.1311 | 1800 | 0.0019 | - |
312
+ | 0.1347 | 1850 | 0.003 | - |
313
+ | 0.1384 | 1900 | 0.003 | - |
314
+ | 0.1420 | 1950 | 0.0024 | - |
315
+ | 0.1456 | 2000 | 0.0023 | - |
316
+ | 0.1493 | 2050 | 0.002 | - |
317
+ | 0.1529 | 2100 | 0.0033 | - |
318
+ | 0.1566 | 2150 | 0.0032 | - |
319
+ | 0.1602 | 2200 | 0.0048 | - |
320
+ | 0.1638 | 2250 | 0.004 | - |
321
+ | 0.1675 | 2300 | 0.0032 | - |
322
+ | 0.1711 | 2350 | 0.0033 | - |
323
+ | 0.1748 | 2400 | 0.0036 | - |
324
+ | 0.1784 | 2450 | 0.0031 | - |
325
+ | 0.1820 | 2500 | 0.0024 | - |
326
+ | 0.1857 | 2550 | 0.0016 | - |
327
+ | 0.1893 | 2600 | 0.0024 | - |
328
+ | 0.1930 | 2650 | 0.0034 | - |
329
+ | 0.1966 | 2700 | 0.0022 | - |
330
+ | 0.2002 | 2750 | 0.0021 | - |
331
+ | 0.2039 | 2800 | 0.0022 | - |
332
+ | 0.2075 | 2850 | 0.0012 | - |
333
+ | 0.2112 | 2900 | 0.0022 | - |
334
+ | 0.2148 | 2950 | 0.001 | - |
335
+ | 0.2185 | 3000 | 0.0007 | - |
336
+ | 0.2221 | 3050 | 0.0011 | - |
337
+ | 0.2257 | 3100 | 0.0008 | - |
338
+ | 0.2294 | 3150 | 0.0008 | - |
339
+ | 0.2330 | 3200 | 0.0016 | - |
340
+ | 0.2367 | 3250 | 0.0026 | - |
341
+ | 0.2403 | 3300 | 0.0018 | - |
342
+ | 0.2439 | 3350 | 0.0021 | - |
343
+ | 0.2476 | 3400 | 0.001 | - |
344
+ | 0.2512 | 3450 | 0.002 | - |
345
+ | 0.2549 | 3500 | 0.0017 | - |
346
+ | 0.2585 | 3550 | 0.0011 | - |
347
+ | 0.2621 | 3600 | 0.0007 | - |
348
+ | 0.2658 | 3650 | 0.0019 | - |
349
+ | 0.2694 | 3700 | 0.0023 | - |
350
+ | 0.2731 | 3750 | 0.0022 | - |
351
+ | 0.2767 | 3800 | 0.0015 | - |
352
+ | 0.2803 | 3850 | 0.0016 | - |
353
+ | 0.2840 | 3900 | 0.0017 | - |
354
+ | 0.2876 | 3950 | 0.0041 | - |
355
+ | 0.2913 | 4000 | 0.0028 | - |
356
+ | 0.2949 | 4050 | 0.0032 | - |
357
+ | 0.2986 | 4100 | 0.004 | - |
358
+ | 0.3022 | 4150 | 0.0025 | - |
359
+ | 0.3058 | 4200 | 0.0026 | - |
360
+ | 0.3095 | 4250 | 0.0024 | - |
361
+ | 0.3131 | 4300 | 0.0015 | - |
362
+ | 0.3168 | 4350 | 0.0013 | - |
363
+ | 0.3204 | 4400 | 0.0026 | - |
364
+ | 0.3240 | 4450 | 0.0017 | - |
365
+ | 0.3277 | 4500 | 0.0015 | - |
366
+ | 0.3313 | 4550 | 0.0013 | - |
367
+ | 0.3350 | 4600 | 0.0012 | - |
368
+ | 0.3386 | 4650 | 0.0009 | - |
369
+ | 0.3422 | 4700 | 0.0008 | - |
370
+ | 0.3459 | 4750 | 0.0009 | - |
371
+ | 0.3495 | 4800 | 0.0005 | - |
372
+ | 0.3532 | 4850 | 0.0005 | - |
373
+ | 0.3568 | 4900 | 0.001 | - |
374
+ | 0.3604 | 4950 | 0.0005 | - |
375
+ | 0.3641 | 5000 | 0.0003 | - |
376
+ | 0.3677 | 5050 | 0.0011 | - |
377
+ | 0.3714 | 5100 | 0.0006 | - |
378
+ | 0.3750 | 5150 | 0.0007 | - |
379
+ | 0.3786 | 5200 | 0.0006 | - |
380
+ | 0.3823 | 5250 | 0.0007 | - |
381
+ | 0.3859 | 5300 | 0.0005 | - |
382
+ | 0.3896 | 5350 | 0.001 | - |
383
+ | 0.3932 | 5400 | 0.0006 | - |
384
+ | 0.3969 | 5450 | 0.0011 | - |
385
+ | 0.4005 | 5500 | 0.0009 | - |
386
+ | 0.4041 | 5550 | 0.001 | - |
387
+ | 0.4078 | 5600 | 0.001 | - |
388
+ | 0.4114 | 5650 | 0.0011 | - |
389
+ | 0.4151 | 5700 | 0.0007 | - |
390
+ | 0.4187 | 5750 | 0.0008 | - |
391
+ | 0.4223 | 5800 | 0.0009 | - |
392
+ | 0.4260 | 5850 | 0.0004 | - |
393
+ | 0.4296 | 5900 | 0.0007 | - |
394
+ | 0.4333 | 5950 | 0.0005 | - |
395
+ | 0.4369 | 6000 | 0.0011 | - |
396
+ | 0.4405 | 6050 | 0.0007 | - |
397
+ | 0.4442 | 6100 | 0.0007 | - |
398
+ | 0.4478 | 6150 | 0.0003 | - |
399
+ | 0.4515 | 6200 | 0.0004 | - |
400
+ | 0.4551 | 6250 | 0.0006 | - |
401
+ | 0.4587 | 6300 | 0.0003 | - |
402
+ | 0.4624 | 6350 | 0.001 | - |
403
+ | 0.4660 | 6400 | 0.0006 | - |
404
+ | 0.4697 | 6450 | 0.0009 | - |
405
+ | 0.4733 | 6500 | 0.0008 | - |
406
+ | 0.4770 | 6550 | 0.0009 | - |
407
+ | 0.4806 | 6600 | 0.0005 | - |
408
+ | 0.4842 | 6650 | 0.0009 | - |
409
+ | 0.4879 | 6700 | 0.0009 | - |
410
+ | 0.4915 | 6750 | 0.0012 | - |
411
+ | 0.4952 | 6800 | 0.0004 | - |
412
+ | 0.4988 | 6850 | 0.0005 | - |
413
+ | 0.5024 | 6900 | 0.0009 | - |
414
+ | 0.5061 | 6950 | 0.0014 | - |
415
+ | 0.5097 | 7000 | 0.0005 | - |
416
+ | 0.5134 | 7050 | 0.0007 | - |
417
+ | 0.5170 | 7100 | 0.0009 | - |
418
+ | 0.5206 | 7150 | 0.0011 | - |
419
+ | 0.5243 | 7200 | 0.001 | - |
420
+ | 0.5279 | 7250 | 0.0021 | - |
421
+ | 0.5316 | 7300 | 0.0015 | - |
422
+ | 0.5352 | 7350 | 0.001 | - |
423
+ | 0.5388 | 7400 | 0.001 | - |
424
+ | 0.5425 | 7450 | 0.0018 | - |
425
+ | 0.5461 | 7500 | 0.0009 | - |
426
+ | 0.5498 | 7550 | 0.0008 | - |
427
+ | 0.5534 | 7600 | 0.0004 | - |
428
+ | 0.5571 | 7650 | 0.0007 | - |
429
+ | 0.5607 | 7700 | 0.0009 | - |
430
+ | 0.5643 | 7750 | 0.0011 | - |
431
+ | 0.5680 | 7800 | 0.0006 | - |
432
+ | 0.5716 | 7850 | 0.0016 | - |
433
+ | 0.5753 | 7900 | 0.0016 | - |
434
+ | 0.5789 | 7950 | 0.0009 | - |
435
+ | 0.5825 | 8000 | 0.0017 | - |
436
+ | 0.5862 | 8050 | 0.0017 | - |
437
+ | 0.5898 | 8100 | 0.001 | - |
438
+ | 0.5935 | 8150 | 0.001 | - |
439
+ | 0.5971 | 8200 | 0.0005 | - |
440
+ | 0.6007 | 8250 | 0.0008 | - |
441
+ | 0.6044 | 8300 | 0.0003 | - |
442
+ | 0.6080 | 8350 | 0.0005 | - |
443
+ | 0.6117 | 8400 | 0.0006 | - |
444
+ | 0.6153 | 8450 | 0.0006 | - |
445
+ | 0.6189 | 8500 | 0.0007 | - |
446
+ | 0.6226 | 8550 | 0.0006 | - |
447
+ | 0.6262 | 8600 | 0.0007 | - |
448
+ | 0.6299 | 8650 | 0.0008 | - |
449
+ | 0.6335 | 8700 | 0.0005 | - |
450
+ | 0.6372 | 8750 | 0.001 | - |
451
+ | 0.6408 | 8800 | 0.0011 | - |
452
+ | 0.6444 | 8850 | 0.0008 | - |
453
+ | 0.6481 | 8900 | 0.0008 | - |
454
+ | 0.6517 | 8950 | 0.0007 | - |
455
+ | 0.6554 | 9000 | 0.0006 | - |
456
+ | 0.6590 | 9050 | 0.0008 | - |
457
+ | 0.6626 | 9100 | 0.0004 | - |
458
+ | 0.6663 | 9150 | 0.0007 | - |
459
+ | 0.6699 | 9200 | 0.0007 | - |
460
+ | 0.6736 | 9250 | 0.0002 | - |
461
+ | 0.6772 | 9300 | 0.0007 | - |
462
+ | 0.6808 | 9350 | 0.0007 | - |
463
+ | 0.6845 | 9400 | 0.0006 | - |
464
+ | 0.6881 | 9450 | 0.0007 | - |
465
+ | 0.6918 | 9500 | 0.0009 | - |
466
+ | 0.6954 | 9550 | 0.0007 | - |
467
+ | 0.6990 | 9600 | 0.0006 | - |
468
+ | 0.7027 | 9650 | 0.0009 | - |
469
+ | 0.7063 | 9700 | 0.0005 | - |
470
+ | 0.7100 | 9750 | 0.0006 | - |
471
+ | 0.7136 | 9800 | 0.001 | - |
472
+ | 0.7173 | 9850 | 0.0004 | - |
473
+ | 0.7209 | 9900 | 0.0006 | - |
474
+ | 0.7245 | 9950 | 0.0006 | - |
475
+ | 0.7282 | 10000 | 0.0003 | - |
476
+ | 0.7318 | 10050 | 0.0009 | - |
477
+ | 0.7355 | 10100 | 0.0006 | - |
478
+ | 0.7391 | 10150 | 0.0011 | - |
479
+ | 0.7427 | 10200 | 0.0008 | - |
480
+ | 0.7464 | 10250 | 0.0006 | - |
481
+ | 0.7500 | 10300 | 0.0008 | - |
482
+ | 0.7537 | 10350 | 0.0006 | - |
483
+ | 0.7573 | 10400 | 0.0005 | - |
484
+ | 0.7609 | 10450 | 0.0008 | - |
485
+ | 0.7646 | 10500 | 0.0007 | - |
486
+ | 0.7682 | 10550 | 0.0005 | - |
487
+ | 0.7719 | 10600 | 0.0007 | - |
488
+ | 0.7755 | 10650 | 0.0011 | - |
489
+ | 0.7791 | 10700 | 0.0011 | - |
490
+ | 0.7828 | 10750 | 0.0008 | - |
491
+ | 0.7864 | 10800 | 0.0003 | - |
492
+ | 0.7901 | 10850 | 0.0006 | - |
493
+ | 0.7937 | 10900 | 0.0009 | - |
494
+ | 0.7973 | 10950 | 0.0006 | - |
495
+ | 0.8010 | 11000 | 0.0008 | - |
496
+ | 0.8046 | 11050 | 0.0005 | - |
497
+ | 0.8083 | 11100 | 0.0014 | - |
498
+ | 0.8119 | 11150 | 0.0007 | - |
499
+ | 0.8156 | 11200 | 0.0004 | - |
500
+ | 0.8192 | 11250 | 0.001 | - |
501
+ | 0.8228 | 11300 | 0.0005 | - |
502
+ | 0.8265 | 11350 | 0.0003 | - |
503
+ | 0.8301 | 11400 | 0.0005 | - |
504
+ | 0.8338 | 11450 | 0.0003 | - |
505
+ | 0.8374 | 11500 | 0.0004 | - |
506
+ | 0.8410 | 11550 | 0.0006 | - |
507
+ | 0.8447 | 11600 | 0.0006 | - |
508
+ | 0.8483 | 11650 | 0.0006 | - |
509
+ | 0.8520 | 11700 | 0.0005 | - |
510
+ | 0.8556 | 11750 | 0.0008 | - |
511
+ | 0.8592 | 11800 | 0.0009 | - |
512
+ | 0.8629 | 11850 | 0.0007 | - |
513
+ | 0.8665 | 11900 | 0.0012 | - |
514
+ | 0.8702 | 11950 | 0.0003 | - |
515
+ | 0.8738 | 12000 | 0.0006 | - |
516
+ | 0.8774 | 12050 | 0.0007 | - |
517
+ | 0.8811 | 12100 | 0.0008 | - |
518
+ | 0.8847 | 12150 | 0.0003 | - |
519
+ | 0.8884 | 12200 | 0.0006 | - |
520
+ | 0.8920 | 12250 | 0.0006 | - |
521
+ | 0.8957 | 12300 | 0.0004 | - |
522
+ | 0.8993 | 12350 | 0.0005 | - |
523
+ | 0.9029 | 12400 | 0.0005 | - |
524
+ | 0.9066 | 12450 | 0.0006 | - |
525
+ | 0.9102 | 12500 | 0.0004 | - |
526
+ | 0.9139 | 12550 | 0.0005 | - |
527
+ | 0.9175 | 12600 | 0.0003 | - |
528
+ | 0.9211 | 12650 | 0.0004 | - |
529
+ | 0.9248 | 12700 | 0.0005 | - |
530
+ | 0.9284 | 12750 | 0.0006 | - |
531
+ | 0.9321 | 12800 | 0.0004 | - |
532
+ | 0.9357 | 12850 | 0.0005 | - |
533
+ | 0.9393 | 12900 | 0.0005 | - |
534
+ | 0.9430 | 12950 | 0.0011 | - |
535
+ | 0.9466 | 13000 | 0.0004 | - |
536
+ | 0.9503 | 13050 | 0.0007 | - |
537
+ | 0.9539 | 13100 | 0.0005 | - |
538
+ | 0.9575 | 13150 | 0.0006 | - |
539
+ | 0.9612 | 13200 | 0.0005 | - |
540
+ | 0.9648 | 13250 | 0.0007 | - |
541
+ | 0.9685 | 13300 | 0.0007 | - |
542
+ | 0.9721 | 13350 | 0.0004 | - |
543
+ | 0.9758 | 13400 | 0.0005 | - |
544
+ | 0.9794 | 13450 | 0.0005 | - |
545
+ | 0.9830 | 13500 | 0.0004 | - |
546
+ | 0.9867 | 13550 | 0.0006 | - |
547
+ | 0.9903 | 13600 | 0.0004 | - |
548
+ | 0.9940 | 13650 | 0.0007 | - |
549
+ | 0.9976 | 13700 | 0.0007 | - |
550
 
551
  ### Framework Versions
552
  - Python: 3.12.11
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:448a09481ac348f2afe99c78065de4c9de571335e5977fdfdfb65993b3ede660
3
  size 546938168
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:457d9b0d8028678927ed4846570853e565ee203fdb2f9c047e05af85db155038
3
  size 546938168
model_head.pkl CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e29762c9d04603d846ab15c84835359d9df24ff62f93e2775c4fe24757666db9
3
  size 388927
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c6ba1f9b176554621cb60c61c54b8008efa143eb5186b8dcb3056f18ee9a5e5c
3
  size 388927