Hugging Face's logo

Michionlion
/
Scan-Plan-LFM2.5-Encoder-230M-v11-FP32

ONNX
onnxruntime
onnxruntime-web
information-extraction
schedule-extraction
Model card Files
xet
 Community
Michionlion commited on
Commit
7ecb3c2
·
verified ·
1 Parent(s): ac0853f

Upload Scan Plan v11 FP32 deployment bundle

Browse files
Files changed (9) hide show
  1. chat_template.jinja +64 -0
  2. config.json +40 -0
  3. label_mapping.json +154 -0
  4. metadata.json +40 -0
  5. model.onnx +3 -0
  6. preprocessing_contract.json +9 -0
  7. schedule-parser-schema-v1.json +196 -0
  8. tokenizer.json +0 -0
  9. tokenizer_config.json +11 -0
chat_template.jinja ADDED
@@ -0,0 +1,64 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {{- bos_token -}}
2
+ {%- set keep_past_thinking = keep_past_thinking | default(false) -%}
3
+ {%- set ns = namespace(system_prompt="") -%}
4
+ {%- if messages[0]["role"] == "system" -%}
5
+ {%- set sys_content = messages[0]["content"] -%}
6
+ {%- if sys_content is not string -%}
7
+ {%- for item in sys_content -%}
8
+ {%- if item["type"] == "text" -%}
9
+ {%- set ns.system_prompt = ns.system_prompt + item["text"] -%}
10
+ {%- endif -%}
11
+ {%- endfor -%}
12
+ {%- else -%}
13
+ {%- set ns.system_prompt = sys_content -%}
14
+ {%- endif -%}
15
+ {%- set messages = messages[1:] -%}
16
+ {%- endif -%}
17
+ {%- if tools -%}
18
+ {%- set ns.system_prompt = ns.system_prompt + ("\n" if ns.system_prompt else "") + "List of tools: [" -%}
19
+ {%- for tool in tools -%}
20
+ {%- if tool is not string -%}
21
+ {%- set tool = tool | tojson -%}
22
+ {%- endif -%}
23
+ {%- set ns.system_prompt = ns.system_prompt + tool -%}
24
+ {%- if not loop.last -%}
25
+ {%- set ns.system_prompt = ns.system_prompt + ", " -%}
26
+ {%- endif -%}
27
+ {%- endfor -%}
28
+ {%- set ns.system_prompt = ns.system_prompt + "]" -%}
29
+ {%- endif -%}
30
+ {%- if ns.system_prompt -%}
31
+ {{- "<|im_start|>system\n" + ns.system_prompt + "<|im_end|>\n" -}}
32
+ {%- endif -%}
33
+ {%- set ns.last_assistant_index = -1 -%}
34
+ {%- for message in messages -%}
35
+ {%- if message["role"] == "assistant" -%}
36
+ {%- set ns.last_assistant_index = loop.index0 -%}
37
+ {%- endif -%}
38
+ {%- endfor -%}
39
+ {%- for message in messages -%}
40
+ {{- "<|im_start|>" + message["role"] + "\n" -}}
41
+ {%- set content = message["content"] -%}
42
+ {%- if content is not string -%}
43
+ {%- set ns.content = "" -%}
44
+ {%- for item in content -%}
45
+ {%- if item["type"] == "image" -%}
46
+ {%- set ns.content = ns.content + "<image>" -%}
47
+ {%- elif item["type"] == "text" -%}
48
+ {%- set ns.content = ns.content + item["text"] -%}
49
+ {%- else -%}
50
+ {%- set ns.content = ns.content + item | tojson -%}
51
+ {%- endif -%}
52
+ {%- endfor -%}
53
+ {%- set content = ns.content -%}
54
+ {%- endif -%}
55
+ {%- if message["role"] == "assistant" and not keep_past_thinking and loop.index0 != ns.last_assistant_index -%}
56
+ {%- if "</think>" in content -%}
57
+ {%- set content = content.split("</think>")[-1] | trim -%}
58
+ {%- endif -%}
59
+ {%- endif -%}
60
+ {{- content + "<|im_end|>\n" -}}
61
+ {%- endfor -%}
62
+ {%- if add_generation_prompt -%}
63
+ {{- "<|im_start|>assistant\n" -}}
64
+ {%- endif -%}
config.json ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "inference": {
3
+ "acceptance_confidence_threshold": 0.7,
4
+ "irrelevant_context_masking": true,
5
+ "issue_threshold": 0.5,
6
+ "preflight_hazard_guard": true,
7
+ "risk_threshold": 0.5,
8
+ "slot_presence_threshold": 0.6,
9
+ "span_decoder": "guarded_hybrid",
10
+ "treatment_control_rescue": false
11
+ },
12
+ "inputs": [
13
+ "input_ids",
14
+ "attention_mask"
15
+ ],
16
+ "max_phases": 4,
17
+ "max_sequence_length": 512,
18
+ "outputs": [
19
+ "token_logits",
20
+ "status_logits",
21
+ "phase_count_logits",
22
+ "reference_logits",
23
+ "issue_logits",
24
+ "risk_logits",
25
+ "window_mode_logits",
26
+ "slot_presence_logits",
27
+ "slot_start_logits",
28
+ "slot_end_logits"
29
+ ],
30
+ "preprocessing": {
31
+ "consensus_policy": "parser-model-consensus-v2",
32
+ "deterministic_parser_input_must_be_normalized": true,
33
+ "model_offsets_restored_to_source_text": true,
34
+ "normalizer": "irrelevant-schedule-context-v1",
35
+ "normalizer_required": true,
36
+ "normalizer_source_sha256": "f2d90dd63df2d60a88039d8c8ca42412b0d34cc2d02f33f47132d1bc6073c07a",
37
+ "preflight_hazard_guard_required": true
38
+ },
39
+ "schema_version": "schedule-parser-schema-v1"
40
+ }
label_mapping.json ADDED
@@ -0,0 +1,154 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "issue_types": [
3
+ "ambiguous_reference_event",
4
+ "approximate_interval",
5
+ "conditional_schedule_branch",
6
+ "conflicting_windows",
7
+ "event_triggered_nonrecurring",
8
+ "explicit_visit_list_not_recurrence",
9
+ "frequency_constraint_not_exact_recurrence",
10
+ "incomplete_window",
11
+ "institutional_schedule_not_specified",
12
+ "interval_range",
13
+ "missing_phase_transition_boundary",
14
+ "missing_recurrence_interval",
15
+ "month_or_year_window_unsupported",
16
+ "multiple_interval_choices",
17
+ "multiple_concurrent_schedules",
18
+ "non_recurring_clinical_discretion",
19
+ "overlapping_phase_boundaries",
20
+ "schedule_deferred",
21
+ "too_many_phases",
22
+ "unsupported_cycle_based_interval"
23
+ ],
24
+ "phase_counts": [
25
+ 0,
26
+ 1,
27
+ 2,
28
+ 3,
29
+ 4
30
+ ],
31
+ "reference_events": [
32
+ "C1D1",
33
+ "REGISTRATION",
34
+ "RANDOMIZATION",
35
+ "TREATMENT_START",
36
+ "ENROLLMENT",
37
+ "OTHER",
38
+ "UNSPECIFIED"
39
+ ],
40
+ "span_slots": [
41
+ "REFERENCE_EVENT",
42
+ "WINDOW_SYMMETRIC_VALUE",
43
+ "WINDOW_BEFORE_VALUE",
44
+ "WINDOW_AFTER_VALUE",
45
+ "WINDOW_UNIT",
46
+ "P1_INTERVAL_VALUE",
47
+ "P1_INTERVAL_UNIT",
48
+ "P1_BOUNDARY_ABSOLUTE_VALUE",
49
+ "P1_BOUNDARY_ABSOLUTE_UNIT",
50
+ "P1_BOUNDARY_DURATION_VALUE",
51
+ "P1_BOUNDARY_DURATION_UNIT",
52
+ "P1_BOUNDARY_INDEFINITE",
53
+ "P2_INTERVAL_VALUE",
54
+ "P2_INTERVAL_UNIT",
55
+ "P2_BOUNDARY_ABSOLUTE_VALUE",
56
+ "P2_BOUNDARY_ABSOLUTE_UNIT",
57
+ "P2_BOUNDARY_DURATION_VALUE",
58
+ "P2_BOUNDARY_DURATION_UNIT",
59
+ "P2_BOUNDARY_INDEFINITE",
60
+ "P3_INTERVAL_VALUE",
61
+ "P3_INTERVAL_UNIT",
62
+ "P3_BOUNDARY_ABSOLUTE_VALUE",
63
+ "P3_BOUNDARY_ABSOLUTE_UNIT",
64
+ "P3_BOUNDARY_DURATION_VALUE",
65
+ "P3_BOUNDARY_DURATION_UNIT",
66
+ "P3_BOUNDARY_INDEFINITE",
67
+ "P4_INTERVAL_VALUE",
68
+ "P4_INTERVAL_UNIT",
69
+ "P4_BOUNDARY_ABSOLUTE_VALUE",
70
+ "P4_BOUNDARY_ABSOLUTE_UNIT",
71
+ "P4_BOUNDARY_DURATION_VALUE",
72
+ "P4_BOUNDARY_DURATION_UNIT",
73
+ "P4_BOUNDARY_INDEFINITE"
74
+ ],
75
+ "statuses": [
76
+ "parsed",
77
+ "ambiguous",
78
+ "unsupported"
79
+ ],
80
+ "token_labels": [
81
+ "O",
82
+ "B-REFERENCE_EVENT",
83
+ "I-REFERENCE_EVENT",
84
+ "B-WINDOW_SYMMETRIC_VALUE",
85
+ "I-WINDOW_SYMMETRIC_VALUE",
86
+ "B-WINDOW_BEFORE_VALUE",
87
+ "I-WINDOW_BEFORE_VALUE",
88
+ "B-WINDOW_AFTER_VALUE",
89
+ "I-WINDOW_AFTER_VALUE",
90
+ "B-WINDOW_UNIT",
91
+ "I-WINDOW_UNIT",
92
+ "B-P1_INTERVAL_VALUE",
93
+ "I-P1_INTERVAL_VALUE",
94
+ "B-P1_INTERVAL_UNIT",
95
+ "I-P1_INTERVAL_UNIT",
96
+ "B-P1_BOUNDARY_ABSOLUTE_VALUE",
97
+ "I-P1_BOUNDARY_ABSOLUTE_VALUE",
98
+ "B-P1_BOUNDARY_ABSOLUTE_UNIT",
99
+ "I-P1_BOUNDARY_ABSOLUTE_UNIT",
100
+ "B-P1_BOUNDARY_DURATION_VALUE",
101
+ "I-P1_BOUNDARY_DURATION_VALUE",
102
+ "B-P1_BOUNDARY_DURATION_UNIT",
103
+ "I-P1_BOUNDARY_DURATION_UNIT",
104
+ "B-P1_BOUNDARY_INDEFINITE",
105
+ "I-P1_BOUNDARY_INDEFINITE",
106
+ "B-P2_INTERVAL_VALUE",
107
+ "I-P2_INTERVAL_VALUE",
108
+ "B-P2_INTERVAL_UNIT",
109
+ "I-P2_INTERVAL_UNIT",
110
+ "B-P2_BOUNDARY_ABSOLUTE_VALUE",
111
+ "I-P2_BOUNDARY_ABSOLUTE_VALUE",
112
+ "B-P2_BOUNDARY_ABSOLUTE_UNIT",
113
+ "I-P2_BOUNDARY_ABSOLUTE_UNIT",
114
+ "B-P2_BOUNDARY_DURATION_VALUE",
115
+ "I-P2_BOUNDARY_DURATION_VALUE",
116
+ "B-P2_BOUNDARY_DURATION_UNIT",
117
+ "I-P2_BOUNDARY_DURATION_UNIT",
118
+ "B-P2_BOUNDARY_INDEFINITE",
119
+ "I-P2_BOUNDARY_INDEFINITE",
120
+ "B-P3_INTERVAL_VALUE",
121
+ "I-P3_INTERVAL_VALUE",
122
+ "B-P3_INTERVAL_UNIT",
123
+ "I-P3_INTERVAL_UNIT",
124
+ "B-P3_BOUNDARY_ABSOLUTE_VALUE",
125
+ "I-P3_BOUNDARY_ABSOLUTE_VALUE",
126
+ "B-P3_BOUNDARY_ABSOLUTE_UNIT",
127
+ "I-P3_BOUNDARY_ABSOLUTE_UNIT",
128
+ "B-P3_BOUNDARY_DURATION_VALUE",
129
+ "I-P3_BOUNDARY_DURATION_VALUE",
130
+ "B-P3_BOUNDARY_DURATION_UNIT",
131
+ "I-P3_BOUNDARY_DURATION_UNIT",
132
+ "B-P3_BOUNDARY_INDEFINITE",
133
+ "I-P3_BOUNDARY_INDEFINITE",
134
+ "B-P4_INTERVAL_VALUE",
135
+ "I-P4_INTERVAL_VALUE",
136
+ "B-P4_INTERVAL_UNIT",
137
+ "I-P4_INTERVAL_UNIT",
138
+ "B-P4_BOUNDARY_ABSOLUTE_VALUE",
139
+ "I-P4_BOUNDARY_ABSOLUTE_VALUE",
140
+ "B-P4_BOUNDARY_ABSOLUTE_UNIT",
141
+ "I-P4_BOUNDARY_ABSOLUTE_UNIT",
142
+ "B-P4_BOUNDARY_DURATION_VALUE",
143
+ "I-P4_BOUNDARY_DURATION_VALUE",
144
+ "B-P4_BOUNDARY_DURATION_UNIT",
145
+ "I-P4_BOUNDARY_DURATION_UNIT",
146
+ "B-P4_BOUNDARY_INDEFINITE",
147
+ "I-P4_BOUNDARY_INDEFINITE"
148
+ ],
149
+ "window_modes": [
150
+ "none",
151
+ "symmetric",
152
+ "asymmetric"
153
+ ]
154
+ }
metadata.json ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "base_model": "LiquidAI/LFM2.5-Encoder-230M",
3
+ "base_model_revision": "0b649ad0c684378b03d4d8304f7577a662ab89bc",
4
+ "benchmark_exact_match": null,
5
+ "evaluation_full_parse_exact_match": 0.99921875,
6
+ "inference": {
7
+ "acceptance_confidence_threshold": 0.7,
8
+ "irrelevant_context_masking": true,
9
+ "issue_threshold": 0.5,
10
+ "preflight_hazard_guard": true,
11
+ "risk_threshold": 0.5,
12
+ "slot_presence_threshold": 0.6,
13
+ "span_decoder": "guarded_hybrid",
14
+ "treatment_control_rescue": false
15
+ },
16
+ "max_sequence_length": 512,
17
+ "model": "PocketTrials Schedule Fallback",
18
+ "onnx_equivalence_cases": 3,
19
+ "onnx_equivalence_max_abs_error": 5.7697296142578125e-05,
20
+ "optional_heads": {
21
+ "issue": true,
22
+ "risk": true,
23
+ "structured_span": true,
24
+ "window_mode": true
25
+ },
26
+ "preprocessing": {
27
+ "consensus_policy": "parser-model-consensus-v2",
28
+ "deterministic_parser_input_must_be_normalized": true,
29
+ "model_offsets_restored_to_source_text": true,
30
+ "normalizer": "irrelevant-schedule-context-v1",
31
+ "normalizer_required": true,
32
+ "normalizer_source_sha256": "f2d90dd63df2d60a88039d8c8ca42412b0d34cc2d02f33f47132d1bc6073c07a",
33
+ "preflight_hazard_guard_required": true
34
+ },
35
+ "production_ready": false,
36
+ "quantization": "fp32",
37
+ "schema_version": "schedule-parser-schema-v1",
38
+ "training_dataset_version": "124a405f3d51ccf01a124fe855659af7bcd7b726cd823f66bac56f336ccf81ba",
39
+ "version": "0.1.0-harness"
40
+ }
model.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:35bb5c1821d3dfa143c760a78699d4d9c4e48e9299134dadac10c7590aefd168
3
+ size 920029323
preprocessing_contract.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "consensus_policy": "parser-model-consensus-v2",
3
+ "deterministic_parser_input_must_be_normalized": true,
4
+ "model_offsets_restored_to_source_text": true,
5
+ "normalizer": "irrelevant-schedule-context-v1",
6
+ "normalizer_required": true,
7
+ "normalizer_source_sha256": "f2d90dd63df2d60a88039d8c8ca42412b0d34cc2d02f33f47132d1bc6073c07a",
8
+ "preflight_hazard_guard_required": true
9
+ }
schedule-parser-schema-v1.json ADDED
@@ -0,0 +1,196 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "$schema": "https://json-schema.org/draft/2020-12/schema",
3
+ "$id": "schedule-parser-schema-v1",
4
+ "title": "PocketTrials fallback schedule parser output",
5
+ "type": "object",
6
+ "additionalProperties": false,
7
+ "required": [
8
+ "schema_version",
9
+ "status",
10
+ "reference_event",
11
+ "phases",
12
+ "window",
13
+ "issues",
14
+ "overall_confidence"
15
+ ],
16
+ "properties": {
17
+ "schema_version": {"const": "schedule-parser-schema-v1"},
18
+ "status": {"enum": ["parsed", "ambiguous", "unsupported"]},
19
+ "reference_event": {"$ref": "#/$defs/inferenceReferenceEvent"},
20
+ "phases": {
21
+ "type": "array",
22
+ "maxItems": 4,
23
+ "items": {"$ref": "#/$defs/inferencePhase"}
24
+ },
25
+ "window": {
26
+ "oneOf": [
27
+ {"type": "null"},
28
+ {"$ref": "#/$defs/inferenceWindow"}
29
+ ]
30
+ },
31
+ "issues": {
32
+ "type": "array",
33
+ "items": {"$ref": "#/$defs/inferenceIssue"}
34
+ },
35
+ "overall_confidence": {"type": "number", "minimum": 0, "maximum": 1}
36
+ },
37
+ "$defs": {
38
+ "unit": {"enum": ["day", "week", "month", "year"]},
39
+ "referenceType": {
40
+ "enum": [
41
+ "C1D1",
42
+ "REGISTRATION",
43
+ "RANDOMIZATION",
44
+ "TREATMENT_START",
45
+ "ENROLLMENT",
46
+ "OTHER",
47
+ "UNSPECIFIED"
48
+ ]
49
+ },
50
+ "expected": {
51
+ "type": "object",
52
+ "additionalProperties": false,
53
+ "required": ["status", "reference_event", "phases", "window", "issues"],
54
+ "properties": {
55
+ "status": {"enum": ["parsed", "ambiguous", "unsupported"]},
56
+ "reference_event": {"$ref": "#/$defs/expectedReferenceEvent"},
57
+ "phases": {
58
+ "type": "array",
59
+ "maxItems": 4,
60
+ "items": {"$ref": "#/$defs/expectedPhase"}
61
+ },
62
+ "window": {
63
+ "oneOf": [
64
+ {"type": "null"},
65
+ {"$ref": "#/$defs/expectedWindow"}
66
+ ]
67
+ },
68
+ "issues": {
69
+ "type": "array",
70
+ "uniqueItems": true,
71
+ "items": {"type": "string", "minLength": 1}
72
+ }
73
+ }
74
+ },
75
+ "expectedReferenceEvent": {
76
+ "type": "object",
77
+ "additionalProperties": false,
78
+ "required": ["type", "source_text"],
79
+ "properties": {
80
+ "type": {"$ref": "#/$defs/referenceType"},
81
+ "source_text": {"type": ["string", "null"]}
82
+ }
83
+ },
84
+ "expectedInterval": {
85
+ "type": "object",
86
+ "additionalProperties": false,
87
+ "required": ["value", "unit", "source_text"],
88
+ "properties": {
89
+ "value": {"type": "number", "exclusiveMinimum": 0},
90
+ "unit": {"$ref": "#/$defs/unit"},
91
+ "source_text": {"type": "string", "minLength": 1}
92
+ }
93
+ },
94
+ "expectedBoundary": {
95
+ "type": "object",
96
+ "additionalProperties": false,
97
+ "required": ["type", "value", "unit", "source_text"],
98
+ "properties": {
99
+ "type": {"enum": ["absolute", "duration", "indefinite"]},
100
+ "value": {"type": ["number", "null"]},
101
+ "unit": {
102
+ "oneOf": [{"$ref": "#/$defs/unit"}, {"type": "null"}]
103
+ },
104
+ "source_text": {"type": ["string", "null"]}
105
+ }
106
+ },
107
+ "expectedPhase": {
108
+ "type": "object",
109
+ "additionalProperties": false,
110
+ "required": ["interval", "boundary"],
111
+ "properties": {
112
+ "interval": {"$ref": "#/$defs/expectedInterval"},
113
+ "boundary": {"$ref": "#/$defs/expectedBoundary"}
114
+ }
115
+ },
116
+ "expectedWindow": {
117
+ "type": "object",
118
+ "additionalProperties": false,
119
+ "required": ["before_days", "after_days", "source_text"],
120
+ "properties": {
121
+ "before_days": {"type": "integer", "minimum": 0},
122
+ "after_days": {"type": "integer", "minimum": 0},
123
+ "source_text": {"type": "string", "minLength": 1}
124
+ }
125
+ },
126
+ "confidence": {"type": "number", "minimum": 0, "maximum": 1},
127
+ "inferenceReferenceEvent": {
128
+ "type": "object",
129
+ "additionalProperties": false,
130
+ "required": ["type", "source_text", "confidence"],
131
+ "properties": {
132
+ "type": {"$ref": "#/$defs/referenceType"},
133
+ "source_text": {"type": ["string", "null"]},
134
+ "confidence": {"$ref": "#/$defs/confidence"}
135
+ }
136
+ },
137
+ "inferenceInterval": {
138
+ "type": "object",
139
+ "additionalProperties": false,
140
+ "required": ["value", "unit", "source_text", "confidence"],
141
+ "properties": {
142
+ "value": {"type": "number", "exclusiveMinimum": 0},
143
+ "unit": {"$ref": "#/$defs/unit"},
144
+ "source_text": {"type": "string"},
145
+ "confidence": {"$ref": "#/$defs/confidence"}
146
+ }
147
+ },
148
+ "inferenceBoundary": {
149
+ "type": "object",
150
+ "additionalProperties": false,
151
+ "required": ["type", "value", "unit", "source_text", "confidence"],
152
+ "properties": {
153
+ "type": {"enum": ["absolute", "duration", "indefinite"]},
154
+ "value": {"type": ["number", "null"]},
155
+ "unit": {
156
+ "oneOf": [{"$ref": "#/$defs/unit"}, {"type": "null"}]
157
+ },
158
+ "source_text": {"type": ["string", "null"]},
159
+ "confidence": {"$ref": "#/$defs/confidence"}
160
+ }
161
+ },
162
+ "inferencePhase": {
163
+ "type": "object",
164
+ "additionalProperties": false,
165
+ "required": ["interval", "boundary", "confidence"],
166
+ "properties": {
167
+ "interval": {"$ref": "#/$defs/inferenceInterval"},
168
+ "boundary": {"$ref": "#/$defs/inferenceBoundary"},
169
+ "confidence": {"$ref": "#/$defs/confidence"}
170
+ }
171
+ },
172
+ "inferenceWindow": {
173
+ "type": "object",
174
+ "additionalProperties": false,
175
+ "required": ["before_days", "after_days", "source_text", "confidence"],
176
+ "properties": {
177
+ "before_days": {"type": "integer", "minimum": 0},
178
+ "after_days": {"type": "integer", "minimum": 0},
179
+ "source_text": {"type": "string"},
180
+ "confidence": {"$ref": "#/$defs/confidence"}
181
+ }
182
+ },
183
+ "inferenceIssue": {
184
+ "type": "object",
185
+ "additionalProperties": false,
186
+ "required": ["type", "message", "source_text", "confidence"],
187
+ "properties": {
188
+ "type": {"type": "string", "minLength": 1},
189
+ "message": {"type": "string", "minLength": 1},
190
+ "source_text": {"type": ["string", "null"]},
191
+ "confidence": {"$ref": "#/$defs/confidence"}
192
+ }
193
+ }
194
+ }
195
+ }
196
+
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "backend": "tokenizers",
3
+ "bos_token": "<|startoftext|>",
4
+ "clean_up_tokenization_spaces": false,
5
+ "eos_token": "<|im_end|>",
6
+ "is_local": true,
7
+ "mask_token": "<|mask|>",
8
+ "model_max_length": 1000000000000000019884624838656,
9
+ "pad_token": "<|pad|>",
10
+ "tokenizer_class": "TokenizersBackend"
11
+ }