Add SetFit model

README.md

@@ -13,45 +13,25 @@ tags:
 - text-classification
 - generated_from_setfit_trainer
 widget:
-- text: weather satellite imagery update every 10 minute cloud top temperature colorized
-    reveal area intensity lower level transparent satellite imagery combine data noaa
-    go east west satellite jma himawari satellite providing full coverage weather
-    event world west coast africa west east coast india tile service update recent
-    image every 10 minute 15 km per pixel resolution infrared ir band detects radiation
-    emitted earth???s surface atmosphere cloud ??·infrared window??? portion spectrum
-    radiation wavelength near 103 micrometer term ??·window??? mean pass atmosphere
-    relatively little absorption gas water vapor useful estimating emitting temperature
-    earth???s surface cloud top major advantage ir band sense energy night imagery
-    available 24 hour day advanced baseline imager abi instrument sample radiance
-    earth sixteen spectral band using several array detector instrument???s focal
-    plane single reflective band abi level 1b radiance product channel 1 6 approximate
-    center wavelength 047 064 0865 1378 161 225 micron respectively digital map outgoing
-    radiance value top atmosphere visible nearinfrared ir band single emissive band
-    abi l1b radiance product channel 7 16 approximate center wavelength 39 6185 695
-    734 85 961 1035 112 123 133 micron respectively digital map outgoing radiance
-    value top atmosphere ir band detector sample compressed packetized downlinked
-    ground station level 0 data conversion calibrated geolocated pixel level 1b radiance
-    data detector sample decompressed radiometrically corrected navigated resampled
-    onto invariant output grid referred abi fixed grid
-- text: pipeline operator conducting risk assessment use ecological usa conjunction
-    pipeline information data identify area may suffer longterm permanent environmentalresource
-    damage event hazardous liquid pipeline accident user data encouraged read carefully
-    technical report cited cross reference section understand limitation ecological
-    usa data dataset comprises unusually sensitive area usa data ecological resource
-    state wyoming accordance pipeline safety law 49 usc section 60109 phmsa required
-    identify area unusually sensitive environmental damage event hazardous liquid
-    pipeline accident interaction various regulatory agency pipeline operator private
-    contractor nonprofit conservation organization general public process developed
-    adopted phmsa identify usa ecological resource process consists identifying set
-    candidate ecological resource using approved data source subjecting candidate
-    set filter criterion determine usa identification usa conducted using standardized
-    data processing step automated gi model resultant usa data applicable current
-    future regulatory requirement specified phmsa including limited pipeline integrity
-    management spill response planning additional information concerning ecological
-    usa please refer document listed cross reference section report
-- text: southern ontario land resource information system solris 20
-- text: toronto employment survey summary table
-- text: cordon data directional traffic count
+- text: temperature salinity profile collected ctd cast nw atlantic limit40 w noaa
+    ship delaware ii noaa ship albatross iv 14 january 1997 30 october 1997 data collected
+    12 cruise multiple program ctd cast primarily made conjunction bongo plankton
+    tow plankton data included
+- text: plume height misr 82420 california fire 2020 multiangle imaging spectroradiometer
+    misr team nasa jet propulsion laboratory california institute technology pasadena
+    california provided map wildfire smoke plume height several wildfire california
+    derived data acquired misr instrument board nasa terra satellite august 24 2020
+    misr carry nine fixed camera view scene different angle period seven minute accounting
+    true motion cloud due wind angular parallax cloud different view used derive height
+    smoke plume data contain plume height information czu lightning complex lnu lightning
+    complex scu lightning complex fire observed misr approximately 1210 pm local time
+    august 24 2020 plume height give indication fire intensity indicates whether smoke
+    impacting air quality groundlevel observation plume height also important input
+    air quality model predict smoke go might affect downwind misr plume height map
+    produced using misr interactive explorer minx software
+- text: municipal land transfer tax revenue summary
+- text: aggregated broccoli production yield
+- text: street furniture bicycle parking
 inference: false
 model-index:
 - name: SetFit with sentence-transformers/paraphrase-mpnet-base-v2
@@ -65,16 +45,16 @@ model-index:
       split: test
     metrics:
     - type: accuracy
-      value: 0.295
+      value: 0.595
       name: Accuracy
     - type: precision
-      value: 0.41697416974169743
+      value: 0.7037037037037037
       name: Precision
     - type: recall
-      value: 0.5044642857142857
+      value: 0.8407079646017699
       name: Recall
     - type: f1
-      value: 0.45656565656565656
+      value: 0.7661290322580645
       name: F1
 ---
 
@@ -110,7 +90,7 @@ The model has been trained using an efficient few-shot learning technique that i
 ### Metrics
 | Label   | Accuracy | Precision | Recall | F1     |
 |:--------|:---------|:----------|:-------|:-------|
-| **all** | 0.295    | 0.4170    | 0.5045 | 0.4566 |
+| **all** | 0.595    | 0.7037    | 0.8407 | 0.7661 |
 
 ## Uses
 
@@ -130,7 +110,7 @@ from setfit import SetFitModel
 # Download from the 🤗 Hub
 model = SetFitModel.from_pretrained("lgd/setfit-multilabel")
 # Run inference
-preds = model("cordon data directional traffic count")
+preds = model("street furniture bicycle parking")
 ```
 
 <!--
@@ -162,7 +142,7 @@ preds = model("cordon data directional traffic count")
 ### Training Set Metrics
 | Training set | Min | Median | Max |
 |:-------------|:----|:-------|:----|
-| Word count   | 1   | 4.55   | 11  |
+| Word count   | 1   | 59.4   | 411 |
 
 ### Training Hyperparameters
 - batch_size: (16, 16)
@@ -185,17 +165,17 @@ preds = model("cordon data directional traffic count")
 ### Training Results
 | Epoch | Step | Training Loss | Validation Loss |
 |:-----:|:----:|:-------------:|:---------------:|
-| 0.002 | 1    | 0.3892        | -               |
-| 0.1   | 50   | 0.2344        | -               |
-| 0.2   | 100  | 0.2476        | -               |
-| 0.3   | 150  | 0.0538        | -               |
-| 0.4   | 200  | 0.0805        | -               |
-| 0.5   | 250  | 0.0974        | -               |
-| 0.6   | 300  | 0.0238        | -               |
-| 0.7   | 350  | 0.025         | -               |
-| 0.8   | 400  | 0.0497        | -               |
-| 0.9   | 450  | 0.0227        | -               |
-| 1.0   | 500  | 0.1179        | -               |
+| 0.002 | 1    | 0.2153        | -               |
+| 0.1   | 50   | 0.201         | -               |
+| 0.2   | 100  | 0.1433        | -               |
+| 0.3   | 150  | 0.0812        | -               |
+| 0.4   | 200  | 0.0866        | -               |
+| 0.5   | 250  | 0.0306        | -               |
+| 0.6   | 300  | 0.1093        | -               |
+| 0.7   | 350  | 0.0647        | -               |
+| 0.8   | 400  | 0.0255        | -               |
+| 0.9   | 450  | 0.0421        | -               |
+| 1.0   | 500  | 0.0366        | -               |
 
 ### Framework Versions
 - Python: 3.10.12

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e41900dea02c1c3a7334b67354e5fee53c628344ac880ade9dfb61bad97cea0a
+oid sha256:73ed78543ca4c52601032949c89bb1c5597c6ded96b32366bb9a919cdabc09e1
 size 437967672

model_head.pkl

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a7977aed629ae22bcdd0b7c348817a6cf8ba615d4b0464851c05e304b392410d
+oid sha256:d04c28ebe8951a2c604c4d8f8f0d6a937ce22a97342163d6c56cfae0d683c6da
 size 26916