44-EFT.WP.Data.ModelCards v1.0 | Chapter 6 Tasks & I/O Modes

Home ／ Docs-Technical WhitePaper ／ 44-EFT.WP.Data.ModelCards v1.0

Chapter 6 Tasks & I/O Modes

I. Chapter Purpose & Scope

II. Terminology & Dependencies

• Terminology source: Comprehensive Template v0.1; this chapter adds task–I/O specific fields only.
• Dependent volumes: Data contract/export Core.DataSpec v1.0; units/dimensions Core.Metrology v1.0; path-dependent expressions Core.Equations v1.1. Wrap inline math in backticks (e.g., f_θ(x), p(y|x,θ)); any division/integral/composite operator must use parentheses; no Chinese in formulas/symbols/definitions.

III. Task Types & the task Field (Normative)

task: "<classification|retrieval|generation|asr|segmentation|detection|timeseries|forecasting|ranking|regression>"

IV. Top-Level io_schema (Normative)

io_schema:

version: "v1.0"

inputs:

- {name:"<string>", shape:"<(…)>", dtype:"<fp32|int8|uint8|…>", range:"<[lo,hi] or N/A>", semantics:"<tokenized|waveform|rgb|…>"}

outputs:

- {name:"<string>", shape:"<(…)>", dtype:"<fp32|int8|…>", range:"<[lo,hi]|[0,1]|N/A>", semantics:"<logits|probs|classes|spans|boxes|…>"}

batching: {mode:"<static|dynamic>", max_batch:<int>}

streaming: {enabled:<bool>, chunk_ms:<int?>, lookahead_ms:<int?>}

constraints:

- {type:"shape_compatible", of:["inputs[0]","outputs[0]"]}

- {type:"range", target:"outputs[probs]", rule:"[0,1] & sum==1±1e-6"}

see:

- "EFT.WP.Core.Metrology v1.0:check_dim"

V. Task-Specific I/O Modes (Normative Examples)

• Image Classification (classification)

task: "classification"

io_schema:

inputs:

- {name:"image", shape:"(H,W,3)", dtype:"uint8", range:"[0,255]", semantics:"rgb"}

outputs:

- {name:"probs", shape:"(K,)", dtype:"float32", range:"[0,1]", semantics:"softmax"}

batching: {mode:"dynamic", max_batch: 128}

constraints:

- {type:"range", target:"outputs[probs]", rule:"[0,1] & sum==1±1e-6"}

• Text Generation (generation)

task: "generation"

io_schema:

inputs:

- {name:"tokens_in", shape:"(T_in,)", dtype:"int32", range:"[0,V)", semantics:"tokenized"}

outputs:

- {name:"tokens_out", shape:"(T_out,)", dtype:"int32", range:"[0,V)", semantics:"tokenized"}

- {name:"logprobs", shape:"(T_out,V)", dtype:"float32", range:"(-∞,0]", semantics:"log-softmax"}

streaming: {enabled:true, chunk_ms: 50, lookahead_ms: 0}

• Automatic Speech Recognition (asr)

task: "asr"

io_schema:

inputs:

- {name:"waveform", shape:"(T,)", dtype:"float32", range:"[-1,1]", semantics:"pcm"}

outputs:

- {name:"text", shape:"()", dtype:"string", range:"N/A", semantics:"utf-8"}

constraints:

- {type:"sampling_rate", target:"inputs[waveform]", rule:"f_samp==16000 Hz"}

• Object Detection (detection)

task: "detection"

io_schema:

inputs:

- {name:"image", shape:"(H,W,3)", dtype:"uint8", range:"[0,255]", semantics:"rgb"}

outputs:

- {name:"boxes", shape:"(N,4)", dtype:"float32", range:"[0,1]", semantics:"xywh_norm"}

- {name:"scores", shape:"(N,)", dtype:"float32", range:"[0,1]", semantics:"objectness"}

- {name:"labels", shape:"(N,)", dtype:"int32", range:"[0,K)", semantics:"class_id"}

constraints:

- {type:"range", target:"outputs[boxes]", rule:"[0,1]"}

• Time-Series Forecasting (timeseries|forecasting)

task: "forecasting"

io_schema:

inputs:

- {name:"series", shape:"(T, C)", dtype:"float32", semantics:"zscore"}

- {name:"time_index", shape:"(T,)", dtype:"int64", semantics:"unix_ms"}

outputs:

- {name:"y_hat", shape:"(H, C)", dtype:"float32", semantics:"forecast"}

- {name:"q_hat", shape:"(H, C, Q)", dtype:"float32", semantics:"quantiles"}

constraints:

- {type:"unit", target:"inputs[series]", rule:"SI; check_dim==true"}

(If physical/time/frequency quantities are involved, units/dimensions are validated by Core.Metrology v1.0 with check_dim=true.)

VI. Multi-Task & Multi-Modal I/O

task: ["classification","retrieval"]

io_schema:

version: "v1.0"

modes:

classification:

inputs: [{name:"image", shape:"(H,W,3)", dtype:"uint8", range:"[0,255]"}]

outputs: [{name:"probs", shape:"(K,)", dtype:"float32", range:"[0,1]", semantics:"softmax"}]

retrieval:

inputs: [{name:"query_emb", shape:"(D,)", dtype:"float32"}]

outputs: [{name:"doc_ids", shape:"(M,)", dtype:"int64"}]

VII. Validation Rules & Lint Constraints (Normative)

lint_rules:

- id: IO.RANGE_PROBS

when: "$.io_schema.outputs[?(@.semantics=='softmax')]"

assert: "range == '[0,1]'"

level: error

- id: IO.SHAPE_NONEMPTY

when: "$.io_schema.inputs[*].shape"

assert: "matches('^\\(') and contains(',')"

level: error

- id: IO.DTYPE_ALLOWED

when: "$.io_schema.inputs[*].dtype"

assert: "in_(['float16','float32','int8','int16','int32','uint8','string'])"

level: error

- id: IO.METROLOGY_CHECKDIM

when: "$.io_schema"

assert: "metrology.units=='SI' and metrology.check_dim==true"

level: error

(The above work with Chapter 15 Schema; violations are blocking.)

VIII. Consistency with Evaluation Protocol & Deployment Interfaces

IX. Metrology & Path Dependence (if applicable)

• T_arr = ( 1 / c_ref ) * ( ∫ n_eff d ell )
• T_arr = ( ∫ ( n_eff / c_ref ) d ell )
  and pass check_dim.

X. Machine-Readable Fragment (Drop-in)

task: "classification"

io_schema:

version: "v1.0"

inputs: [{name:"image", shape:"(H,W,3)", dtype:"uint8", range:"[0,255]", semantics:"rgb"}]

outputs: [{name:"probs", shape:"(K,)", dtype:"float32", range:"[0,1]", semantics:"softmax"}]

batching: {mode:"dynamic", max_batch: 128}

streaming: {enabled:false}

see:

- "EFT.WP.Core.Metrology v1.0:check_dim"

XI. Chapter Compliance Checklist

• task and io_schema defined and consistent with evaluation.protocol and deployment interfaces; named sub-modes provided for multi-task settings.
• io_schema specifies shape/dtype/range/semantics, batching/streaming; physical-quantity fields pass check_dim.
• Softmax outputs meet range & normalization constraints; all math/symbols use backticks and parentheses with no Chinese.
• For path-dependent quantities (e.g., T_arr), delta_form/path/measure registered and metrology checks complete.