SliceX AI Predictor

• Binary Text Classification
  • postCustomer Review Sentiment
  • postToxicity Detection
• Multi-Class Text Classification
  • postNews Classification
  • postSentiment Analysis
  • postFinancial Customer Support
  • postFinancial Sentiment Analysis
  • postConversational Language Understanding (En)
  • postConversational Language Understanding (Jp)
• Multi-Label Text Classification
  • postEmotion Detection
• Sequence Labeling
  • postNamed Entity Recognition
• Question Answering
  • postResearch Papers Search
• Custom Trained Model
  • postUser Model Inference

Documentation Powered by Redocly

SliceX AI Predictor API (1.0)

Download OpenAPI specification:Download

Binary Text Classification

Inference with 2 possible classes (e.g. POSITIVE and NEGATIVE).

Customer Review Sentiment

header Parameters

x-api-key required

string (x-api-key)

Request Body schema: application/json

query

string (The string to run the inference on.)

Responses

https://api.slicex.ai/predictor/language/pre-trained/customer-review

Request samples

• Payload
• cURL
• Python3 (requests)
• JavaScript (axios)

Content type

application/json

{

• "query": "This movie was great"

}

Response samples

• 200
• 422

Content type

application/json

{

• "data": {

  • "labels": [

    • "POSITIVE",
    • "NEGATIVE"

    ],
  • "scores": [

    • 0.95,
    • 0.05

    ],
  • "tokens": [

    • "New York"

    ],
  • "token_start_positions": [

    • 2

    ]

  },
• "metadata": {

  • "model_inference_time_ms": 7.8

  }

}

Toxicity Detection

header Parameters

x-api-key required

string (x-api-key)

Request Body schema: application/json

query

string (The string to run the inference on.)

Responses

https://api.slicex.ai/predictor/language/pre-trained/toxicity-detection

Request samples

• Payload
• cURL
• Python3 (requests)
• JavaScript (axios)

Content type

application/json

{

• "query": "This movie was great"

}

Response samples

• 200
• 422

Content type

application/json

{

• "data": {

  • "labels": [

    • "POSITIVE",
    • "NEGATIVE"

    ],
  • "scores": [

    • 0.95,
    • 0.05

    ],
  • "tokens": [

    • "New York"

    ],
  • "token_start_positions": [

    • 2

    ]

  },
• "metadata": {

  • "model_inference_time_ms": 7.8

  }

}

Multi-Class Text Classification

Inference with multiple possible classes. Scores sum up to 1.

News Classification

header Parameters

x-api-key required

string (x-api-key)

Request Body schema: application/json

query

string (The string to run the inference on.)

Responses

https://api.slicex.ai/predictor/language/pre-trained/news-classification

Request samples

• Payload
• cURL
• Python3 (requests)
• JavaScript (axios)

Content type

application/json

{

• "query": "This movie was great"

}

Response samples

• 200
• 422

Content type

application/json

{

• "data": {

  • "labels": [

    • "POSITIVE",
    • "NEGATIVE"

    ],
  • "scores": [

    • 0.95,
    • 0.05

    ],
  • "tokens": [

    • "New York"

    ],
  • "token_start_positions": [

    • 2

    ]

  },
• "metadata": {

  • "model_inference_time_ms": 7.8

  }

}

Sentiment Analysis

header Parameters

x-api-key required

string (x-api-key)

Request Body schema: application/json

query

string (The string to run the inference on.)

Responses

https://api.slicex.ai/predictor/language/pre-trained/sentiment-classification

Request samples

• Payload
• cURL
• Python3 (requests)
• JavaScript (axios)

Content type

application/json

{

• "query": "This movie was great"

}

Response samples

• 200
• 422

Content type

application/json

{

• "data": {

  • "labels": [

    • "POSITIVE",
    • "NEGATIVE"

    ],
  • "scores": [

    • 0.95,
    • 0.05

    ],
  • "tokens": [

    • "New York"

    ],
  • "token_start_positions": [

    • 2

    ]

  },
• "metadata": {

  • "model_inference_time_ms": 7.8

  }

}

Financial Customer Support

header Parameters

x-api-key required

string (x-api-key)

Request Body schema: application/json

query

string (The string to run the inference on.)

Responses

https://api.slicex.ai/predictor/language/pre-trained/financial-customer-support

Request samples

• Payload
• cURL
• Python3 (requests)
• JavaScript (axios)

Content type

application/json

{

• "query": "This movie was great"

}

Response samples

• 200
• 422

Content type

application/json

{

• "data": {

  • "labels": [

    • "POSITIVE",
    • "NEGATIVE"

    ],
  • "scores": [

    • 0.95,
    • 0.05

    ],
  • "tokens": [

    • "New York"

    ],
  • "token_start_positions": [

    • 2

    ]

  },
• "metadata": {

  • "model_inference_time_ms": 7.8

  }

}

Financial Sentiment Analysis

header Parameters

x-api-key required

string (x-api-key)

Request Body schema: application/json

query

string (The string to run the inference on.)

Responses

https://api.slicex.ai/predictor/language/pre-trained/financial-sentiment-classification

Request samples

• Payload
• cURL
• Python3 (requests)
• JavaScript (axios)

Content type

application/json

{

• "query": "This movie was great"

}

Response samples

• 200
• 422

Content type

application/json

{

• "data": {

  • "labels": [

    • "POSITIVE",
    • "NEGATIVE"

    ],
  • "scores": [

    • 0.95,
    • 0.05

    ],
  • "tokens": [

    • "New York"

    ],
  • "token_start_positions": [

    • 2

    ]

  },
• "metadata": {

  • "model_inference_time_ms": 7.8

  }

}

Conversational Language Understanding (En)

header Parameters

x-api-key required

string (x-api-key)

Request Body schema: application/json

query

string (The string to run the inference on.)

Responses

https://api.slicex.ai/predictor/language/pre-trained/conversational-language-understanding-en

Request samples

• Payload
• cURL
• Python3 (requests)
• JavaScript (axios)

Content type

application/json

{

• "query": "This movie was great"

}

Response samples

• 200
• 422

Content type

application/json

{

• "data": {

  • "labels": [

    • "POSITIVE",
    • "NEGATIVE"

    ],
  • "scores": [

    • 0.95,
    • 0.05

    ],
  • "tokens": [

    • "New York"

    ],
  • "token_start_positions": [

    • 2

    ]

  },
• "metadata": {

  • "model_inference_time_ms": 7.8

  }

}

Conversational Language Understanding (Jp)

header Parameters

x-api-key required

string (x-api-key)

Request Body schema: application/json

query

string (The string to run the inference on.)

Responses

https://api.slicex.ai/predictor/language/pre-trained/conversational-language-understanding-jp

Request samples

• Payload
• cURL
• Python3 (requests)
• JavaScript (axios)

Content type

application/json

{

• "query": "This movie was great"

}

Response samples

• 200
• 422

Content type

application/json

{

• "data": {

  • "labels": [

    • "POSITIVE",
    • "NEGATIVE"

    ],
  • "scores": [

    • 0.95,
    • 0.05

    ],
  • "tokens": [

    • "New York"

    ],
  • "token_start_positions": [

    • 2

    ]

  },
• "metadata": {

  • "model_inference_time_ms": 7.8

  }

}

Multi-Label Text Classification

Inference with multiple possible classes. Scores don't sum up to 1.

Emotion Detection

header Parameters

x-api-key required

string (x-api-key)

Request Body schema: application/json

query

string (The string to run the inference on.)

Responses

https://api.slicex.ai/predictor/language/pre-trained/emotion-detection

Request samples

• Payload
• cURL
• Python3 (requests)
• JavaScript (axios)

Content type

application/json

{

• "query": "This movie was great"

}

Response samples

• 200
• 422

Content type

application/json

{

• "data": {

  • "labels": [

    • "POSITIVE",
    • "NEGATIVE"

    ],
  • "scores": [

    • 0.95,
    • 0.05

    ],
  • "tokens": [

    • "New York"

    ],
  • "token_start_positions": [

    • 2

    ]

  },
• "metadata": {

  • "model_inference_time_ms": 7.8

  }

}

Sequence Labeling

Named Entity Recognition. Finds the tokens (e.g. person, location, etc.) in the query.

Named Entity Recognition

header Parameters

x-api-key required

string (x-api-key)

Request Body schema: application/json

query

string (The string to run the inference on.)

Responses

https://api.slicex.ai/predictor/language/pre-trained/named-entity-recognition

Request samples

• Payload
• cURL
• Python3 (requests)
• JavaScript (axios)

Content type

application/json

{

• "query": "This movie was great"

}

Response samples

• 200
• 422

Content type

application/json

{

• "data": {

  • "labels": [

    • "POSITIVE",
    • "NEGATIVE"

    ],
  • "scores": [

    • 0.95,
    • 0.05

    ],
  • "tokens": [

    • "New York"

    ],
  • "token_start_positions": [

    • 2

    ]

  },
• "metadata": {

  • "model_inference_time_ms": 7.8

  }

}

Question Answering

Looks to answer questions based on text passages.

Research Papers Search

header Parameters

x-api-key required

string (x-api-key)

Request Body schema: application/json

query

string (The string to run the inference on.)

Responses

https://api.slicex.ai/predictor/language/pre-trained/research-papers-search

Request samples

• Payload
• cURL
• Python3 (requests)
• JavaScript (axios)

Content type

application/json

{

• "query": "This movie was great"

}

Response samples

• 200
• 422

Content type

application/json

{

• "data": {

  • "labels": [

    • "POSITIVE",
    • "NEGATIVE"

    ],
  • "scores": [

    • 0.95,
    • 0.05

    ],
  • "tokens": [

    • "New York"

    ],
  • "token_start_positions": [

    • 2

    ]

  },
• "metadata": {

  • "model_inference_time_ms": 7.8

  }

}

Custom Trained Model

User Model Inference

path Parameters

model_id required

string (Model Id)

header Parameters

x-api-key required

string (x-api-key)

Request Body schema: application/json

query

string (The string to run the inference on.)

Responses

https://api.slicex.ai/predictor/language/model/{model_id}

Request samples

• Payload
• cURL
• Python3 (requests)
• JavaScript (axios)

Content type

application/json

{

• "query": "This movie was great"

}

Response samples

• 200
• 422

Content type

application/json

{

• "data": {

  • "labels": [

    • "POSITIVE",
    • "NEGATIVE"

    ],
  • "scores": [

    • 0.95,
    • 0.05

    ],
  • "tokens": [

    • "New York"

    ],
  • "token_start_positions": [

    • 2

    ]

  },
• "metadata": {

  • "model_inference_time_ms": 7.8

  }

}